AI Mindset · Model Cheatsheets

Claude

The Senior Colleague Who Tells You When It Is Not Sure

Claude Opus 5.5 changes the tiering math, and most buyers haven’t noticed yet. It landed September 22, 2026 at $4 / $20 per million tokens. That’s 20 percent under Opus 5, and Anthropic says it performs at Fable 5.1 level on most work. It isn’t a Covered Model, so zero data retention is available on it. Regulated traffic that can’t sit under a 30-day window now gets near-top-tier capability without the retention gate. Claude Sonnet 5.5 followed six days later at $2 / $10. Fable 5.1 is still the top tier at $10 / $50, still retention-required, still fixable with one workspace switch. And Sonnet 4.5 now has a firm shutdown date: November 30, 2026.

Verified October 1, 2026Opus 5.5 landed Sept 22, 2026 at $4 / $20Sonnet 5.5 landed Sept 28, 2026 at $2 / $10Opus 5.5 is ZDR-available and cheaper than Opus 5Sonnet 4.5 retires November 30, 202630-day retention on the four Covered Models, with a per-workspace overrideCode · Cowork · Chrome · Tag · Design

1 / Meet Claude

The colleague who will not pretend the work is done

Claude’s best trait isn’t speed. It’s judgment. It will name the evidence that’s missing, argue with your brief, and keep the doubt in plain sight while it still gets the work done.

Personality

Careful without being passive

Claude reads closely and keeps the nuance. Offered a tidy conclusion it can’t support, or a caveat, it takes the caveat.

Strong source fidelity
Structured, professional writing
Useful self-review and error detection

Deploy it for

Work where judgment matters

Long documents. Complex coding. Regulated analysis, technical reviews, and anything that has to show its working months later.

Board and client materials
Repository-scale engineering
Financial, legal and scientific workflows

Choose Claude when

The brief may itself be wrong

Claude earns its keep when the input is thin, argues with itself, or would tempt a model into a polished fiction.

Ask it to inspect the premise
Invite direct disagreement
Require uncertainty to remain visible

Brief, then challenge

Analyze this proposal for the executive team. Before recommending anything, identify what is missing, internally inconsistent or too weakly evidenced to support a decision.

Claude pushes back harder when you say out loud that pushing back is allowed.

Truth-preserving synthesis

Create the strongest defensible synthesis from these sources. Separate: verified facts, reasonable inferences, unresolved questions and claims the material cannot support.

That keeps the doubt useful instead of burying it in a disclaimer.

2 / What Changed

Two new models in six days, a refusal bill that came back, and a shutdown date 60 days out

Two flagship models shipped six days apart, and the cheaper one is the story. Claude Opus 5.5 arrived September 22, 2026 at $4 / $20 per million tokens. Anthropic says it runs at Fable 5.1 level on most work, costs 40 percent less than Opus 5 and writes output more than 30 percent faster. Vendor claims, all three. Claude Sonnet 5.5 followed on September 28 at $2 / $10. Neither is a Covered Model, so neither carries the 30-day retention requirement. Two quieter changes cost money. Refusals that arrive before any output are billed again from September 24. And Sonnet 4.5 stops serving on November 30, 2026.

The nine days that moved the most

September 22 to September 30 did more to a Claude procurement file than the three months before it. Two models launched: Opus 5.5 on September 22 at $4 / $20, Sonnet 5.5 on September 28 at $2 / $10. Neither is a Covered Model, so the retention gate no longer sits between a buyer and a capable model. Sonnet 5 and Opus 5 both moved under “Legacy models” on the pricing page while staying Active in the lifecycle. That makes six models in that split, not two. A tentative floor turned into a firm retirement: Sonnet 4.5, deprecated September 30, gone November 30. Three smaller changes have real teeth. Pre-output refusals in three safety categories are billed again from September 24. The Compliance API Activity Feed stopped returning file names, document names and artifact titles, retroactively, on the same day. And cache diagnostics left beta on September 23, which is the tool you need now that cache reads range from $1 down to $0.20 across the lineup. The action that follows is narrow. Open the routing rule that sends regulated traffic to Opus 5. Open the Claude Code model allowlist. And open the two or three sentences about retention already sitting in your procurement file. All three were written for a lineup that no longer exists.

A refusal that used to be free is now a line item

Check your bill for refusals. From September 24, 2026, Anthropic bills again for refusals that arrive before any output. Three categories are covered, read from stop_details.category: bio, frontier_llm and reasoning_extraction. A billed refusal costs what any other request costs, at the rates of the model that ran it. And it applies on every platform, not just the Claude API. So if your traffic brushes those categories at any volume, a line that used to cost nothing is now on the invoice. Count them first. Then decide whether that traffic should reach the model at all.
· Anthropic’s wording: refusals billed under this change “are charged like any other request, at the rates of the model that ran it,” and the change “applies on all platforms.”
· The item sits on the rolling API release notes under September 24, 2026, so confirm it there before a forecast depends on it.

$4 / $20

Claude Opus 5.5 per MTok, released September 22, 2026: 20 percent under Opus 5, cache hit $0.20, 1M token context, 128k maximum output and no data-retention requirement

$2 / $10

Claude Sonnet 5.5 per MTok, released September 28, 2026: cache hit $0.20, 1M token context, 128k maximum output and no data-retention requirement

Nov 30

Firm retirement date for claude-sonnet-4-5-20250929, deprecated September 30, 2026. Sixty days from October 1, 2026, with Sonnet 5.5 named as the migration target

$0.25

Cache hit per million tokens on Fable 5.1 and Mythos 5.1: 0.025x base input. Opus 5.5 and Sonnet 5.5 both read cached context at $0.20

30 days

Minimum retention on the four Covered Models: Fable 5.1, Mythos 5.1, Fable 5 and Mythos 5. Anthropic’s wording is “at least 30 days,” with a per-workspace override for zero data retention organizations

EFS

Enterprise Frontier Safeguards, announced September 1, 2026: privacy Anthropic equates with zero data retention, with customer-hosted data, rolling out in phases starting later this fall and with no dated milestone published

50%

Batch discount on both input and output: Fable 5.1 at $5 / $25, Opus 5.5 at $2 / $10, Opus 5 at $2.50 / $12.50, Sonnet 5.5 and Sonnet 5 at $1 / $5 and Haiku 4.5 at $0.50 / $2.50

400

The status code returned when a request uses tool_choice type any or tool on Fable 5.1, Mythos 5.1, Opus 5.5 or Sonnet 5.5. Four models now, not two

Claude Opus 5.5 is cheaper than Opus 5 and carries no retention requirement

This is the model that moves a budget and a compliance plan in the same week. Opus 5.5 shipped September 22, 2026 as claude-opus-5-5, at $4 / $20 per million tokens. Opus 5 costs $5 / $25. So the newer model is 20 percent cheaper on the sticker. Anthropic goes further. It says Opus 5.5 costs 40 percent less to run, writes output more than 30 percent faster and performs at Fable 5.1 level on most work. Those are vendor claims. Time them on your own traffic. The sentence a compliance reviewer needs is shorter: Opus 5.5 isn’t a Covered Model, and Anthropic says it’s available with zero data retention. The specification is 1M token context, 128k maximum output, and 300k maximum output on the Batch API in beta. Thinking is always on and adaptive, with a default effort of medium. Fast mode is a research preview on the Claude API. It runs on the Claude API, Amazon Bedrock, the Claude Platform on AWS, Google Cloud and Microsoft Foundry. And it’s on Pro, Max, Team and seat-based Enterprise with increased five-hour usage limits. Three breaking changes ride with it, and the Opus 5.5 card further down spells them out.
· Three first-party pages say Opus 5.5 is available with zero data retention: the launch announcement, the Covered Models article and the retention documentation. One reading of the rolling release-notes page appeared to say otherwise for the September 22 entry. Confirm it on that page before a control description depends on it.

Claude Sonnet 5.5 is the same price as Sonnet 5 and breaks Sonnet 5 code in five ways

Same price, different behavior. Sonnet 5.5 shipped September 28, 2026 as claude-sonnet-5-5, at $2 / $10 per million tokens with a $0.20 cache hit, 1M token context and 128k maximum output. Anthropic says it’s more than 30 percent faster than Sonnet 5 and costs up to 30 percent less for most work. Vendor claims again. One default catches teams out: effort is medium in the Claude apps and high on the Claude Platform. So the same prompt costs more through the API than it does in the app. Five changes will break working Sonnet 5 code. Three of them matter most. Forced tool use returns a 400. Disabling thinking at high effort or below now uses thinking: {"type": "between_tools"} where you used to send "disabled". And the advisor tool rejects Claude Opus 4.8, Claude Opus 4.7 and Claude Sonnet 5 as advisors, which takes out an agent architecture rather than a single call. Sonnet 5.5 isn’t a Covered Model, and Anthropic says zero data retention is available across all platforms.

Breaking change: tool_choice Four models return 400

This is the breaking change most likely to be sitting in your production code right now. Forcing a tool call has been the standard way to guarantee a parseable response. On Fable 5.1, Mythos 5.1, Opus 5.5 and Sonnet 5.5, the tool_choice types any and tool are not supported and return a 400. The types auto and none are unchanged. Anthropic’s stated alternatives are strict tool use or structured outputs. And here’s the trap. A codebase sweep scoped to “the 5.1 models” was correct in September and is wrong now. Scope it to the model ids you actually route to, then search for the two removed values.

any and tool return 400 on Fable 5.1, Mythos 5.1, Opus 5.5 and Sonnet 5.5
auto and none behave as before
Use strict tool use or structured outputs instead
A sweep scoped to “5.1 models” now misses two more
This fails the request; it does not degrade quietly

Breaking change: thinking blocks 5.1 returns 400

A team can split down the middle on identical code, and the only difference is how old their accounts are. Here’s why. Thinking blocks are preserved only for the model that produced them, or a newer one. For accounts created on or after August 31, 2026, replaying a thinking block after a changed system prompt, changed tools or a changed message returns a 400. The beta header thinking-binding-controls-2026-08-01 controls that behavior. The exposure is any pattern that caches an assistant turn and replays it against an edited prompt. That’s common in agent frameworks and in evaluation harnesses that mutate the system prompt between runs. An older account may never see the error. So one engineer ships and another one can’t.

Thinking blocks survive only into the producing model or newer
Accounts created on or after August 31, 2026 get a 400 on a changed system prompt, tools or message
Controlled by the beta header thinking-binding-controls-2026-08-01
Older and newer accounts can behave differently on the same code

Thinking blocks are now account-bound, and this one fails silently

The dangerous version of a breaking change is the one that returns a 200. On Sonnet 5.5, a thinking block works only in the account that produced it, or in an account linked to it. Send one from a different account and the API drops the block before the model ever sees it. The request succeeds. You get an answer. The reasoning you paid to generate and carefully replayed was thrown away, and nothing in the response says so. Read that beside the other thinking-block rule, because the failure modes are opposites. The account-creation-date rule returns a clean 400 you can catch in a test. This one returns a quiet, slightly worse answer. Anywhere you share cached assistant turns across accounts, an evaluation harness, a shared fixture library, a multi-tenant agent, you can’t detect this from status codes. Compare outputs instead.

Sonnet 5.5 thinking blocks work only in the producing account or a linked one
A block from another account is dropped and the request still returns 200
The opposite failure mode from the 400 on a changed prompt, tools or message
Shared fixtures and multi-tenant agents are the exposure
Detect it by comparing output quality, not by watching for errors

A floor just became a date: Sonnet 4.5 stops serving on November 30, 2026

The floor got exercised. That’s the single most useful thing on the deprecations page today. And it’s a lesson about floors generally. claude-sonnet-4-5-20250929 carried a tentative line reading “Not sooner than September 29, 2026” for months. The day after that floor expired, Anthropic deprecated the model. Deprecated September 30, 2026, with retirement on the Claude API scheduled for November 30, 2026, and a migration pointer to Claude Sonnet 5.5. So you have 60 days from October 1, and the work is a model-id change plus five breaking changes. Start now rather than in November. Two Active snapshots still carry floors inside that same window. claude-haiku-4-5-20251001: “Not sooner than October 15, 2026,” which is 14 days out with no deprecation notice attached. claude-opus-4-5-20251101: “Not sooner than November 24, 2026,” 54 days out. Treat each as the earliest date you could be forced to move, and now you know Anthropic does act on them. Further out, every floor sits in 2027. The table below lists them, and all of those models are Active. Every model in the lineup now has a published floor, including both Mythos models. The one deprecated Mythos line is Claude Mythos Preview, deprecated June 9, 2026, with a retirement date still to be announced. The deprecations page is a living reference, not a dated article. Diarize the date, re-read the source page on the same cadence, and keep a dated copy of the row you depend on. A date written down once and trusted is how you miss a deadline that moved.

Model	Status on October 1, 2026	Retirement as published
Claude Sonnet 4.5, `claude-sonnet-4-5-20250929`	Deprecated September 30, 2026	November 30, 2026
Claude Haiku 4.5, `claude-haiku-4-5-20251001`	Active	Not sooner than October 15, 2026
Claude Opus 4.5, `claude-opus-4-5-20251101`	Active	Not sooner than November 24, 2026
Claude Opus 4.8	Active	Not sooner than May 28, 2027
Claude Fable 5 and Claude Mythos 5	Active	Not sooner than June 9, 2027
Claude Sonnet 5	Active	Not sooner than June 30, 2027
Claude Opus 5	Active	Not sooner than July 24, 2027
Claude Fable 5.1 and Claude Mythos 5.1	Active; Mythos 5.1 invite only	Not sooner than September 1, 2027
Claude Opus 5.5	Active	Not sooner than September 22, 2027
Claude Sonnet 5.5	Active	Not sooner than September 28, 2027
Claude Mythos Preview	Deprecated June 9, 2026	To be announced

The 30-day retention rule, and the workspace switch that gets you past it

Someone in your organization has already written off Fable 5.1. Zero data retention shop, retention-required model, conversation over. There’s a switch. Most people don’t know it exists. Here’s the order to read it in. On Anthropic’s direct terms, Claude Fable 5.1 and Claude Mythos 5.1 do require 30-day data retention. The fix is a configuration flag Anthropic documents itself. An organization with a zero data retention arrangement can turn on 30-day retention for one workspace and leave every other workspace at zero. The path is self-serve. In Claude Console, open Settings, then Workspaces, pick the workspace, open its Privacy controls tab, and turn on 30-day data retention. It settles the mixed-estate question too. Some traffic on 5.1 under retention, some on Opus 5.5 without it: that needs a routing rule a compliance reviewer can read, not one buried in application code. The workspace boundary is that rule, and it is Anthropic’s, so you don’t have to invent one. The failure mode is readable as well. A request from an organization or workspace without retention enabled comes back as a 400 invalid_request_error. Everything here describes the direct Anthropic route. What Microsoft publishes for the same model as a subprocessor is a separate question, and it gets its own callout further down this section.
· Anthropic’s wording: “Organizations with a ZDR arrangement can make these models available in a specific workspace by enabling 30-day retention for that workspace only. Other workspaces in the organization keep zero data retention.”
· The 400 message: “In order to access this model, your organization or workspace must have data retention enabled.”
· Anthropic’s September 1 API release note still says both models require 30-day retention and aren’t available under zero data retention unless expressly authorized by Anthropic.
· The retention page is a living document with no publication date. Read it yourself before any of this goes into a control description.

What “deleted after 30 days” actually says, and what a Covered Model is

“Deleted after 30 days” promises more than Anthropic does. The wording is at least 30 days, not exactly 30. Then two carve-outs: a safety investigation, and a legal requirement to keep the data. Neither carve-out carries a duration in that sentence. The durations a privacy review actually stops on are printed somewhere else, and they’re long. Anthropic now prints the big one itself. If a chat or session is flagged, Anthropic may retain inputs and outputs for up to two years. Three more tails sit on the same page and almost nobody has them in a data protection impact assessment. The Activity Feed retains data for six years. Local session transcripts are stored for six years by default. Remote session transcripts, meaning Cowork in the cloud, are retained for six years unless a user deletes the session sooner. Only one long duration is still not Anthropic’s own words: trust and safety classification scores kept for up to seven years. That figure appears on Microsoft’s page describing Anthropic and on no Anthropic page, so confirm it before a privacy review leans on it. Two more things belong in the same breath. First, what a Covered Model is. In plain words: a model so much stronger than the last generation that misuse becomes a real risk. Four are designated: Mythos 5.1 and Fable 5.1, both on August 31, 2026, and Mythos 5 and Fable 5, both on June 9, 2026. Second, retained data doesn’t always sit in the same place. On the Claude API, including the Claude Platform on AWS, Anthropic handles it. On Amazon Bedrock and the Gemini Enterprise Agent Platform, it stays inside your cloud provider’s environment. That’s the data-residency answer, and it changes with the platform rather than being uniform.
· Anthropic’s definition of a Covered Model: a Claude model whose capabilities, for example in software engineering, agentic workflows, scientific reasoning or cybersecurity, are a substantial step up from prior generations and create elevated risk if misused.
· Anthropic’s wording: prompts and model completions “are retained for at least 30 days and then automatically deleted, unless they are subject to a safety investigation or we are legally required to maintain them.”
· The two-year figure and the three six-year tails: Anthropic’s own API and data retention documentation, a living page with no date line. The seven-year classification-score figure: “Anthropic models in Microsoft Online Services,” Microsoft Learn, a living page stamped September 18, 2026 and read October 1, 2026.
· The Covered Models article carries no publication date, only a relative “last updated” line, so it can anchor content but not a date. Confirm it directly.

Enterprise Frontier Safeguards and the interim zero data retention offer

Anthropic dropped this on launch day, which is the main reason nobody read it. Enterprise Frontier Safeguards, or EFS, is described as giving customers complete privacy, and Anthropic explicitly equates that with a zero data retention policy. The mechanism is plain: customers store their data on their own cloud infrastructure instead of on Anthropic’s systems. The scope it names: Claude Code, Claude Enterprise, the Claude Platform, Amazon Bedrock, the Claude Platform on AWS, the Gemini Enterprise Agent Platform and Microsoft Foundry. The schedule is where it gets slippery. The September 1 article now says the rollout happens in phases starting later this fall, which is future tense and vaguer than it was. The request form says EFS is rolling out to eligible customers in phases, with the goal of making it broadly available this fall, which is present tense. Two first-party pages, two tenses, and no named phase, no named customer, no dated milestone on either. Nothing published between September 17 and October 1 says a phase shipped. So treat EFS as announced and unscheduled. In the meantime, Anthropic says that until EFS is available, eligible customers will be able to use Fable 5.1 with zero data retention. The dedicated September 1 article extends that to Fable 5 as well and describes the arrangement as available for a limited time. The same article says where the policy came from: Anthropic introduced 30-day data retention starting with Fable 5. Treat the interim offer as something you ask your account team for by name, with a date attached. Not as a benefit you already hold.
· Both items were published by Anthropic on September 1, 2026, the same day as the 5.1 launch.

Where a blocked request goes Two categories, not one

If a runbook in your product names the fallback model, somebody guessed. No first-party page read on October 1, 2026 supports that precision. Anthropic’s readable wording is plural and unnamed: “Our safeguards do, however, still redirect several kinds of dual-use cybersecurity tasks...to our Opus models. This includes penetration testing, exploit generation, and binary-based vulnerability scanning.” So write “an Opus model” in runbooks and user-facing notes, not a version number. There’s a second routing category on the same page. It’s easy to read past: “Queries related to research and development in the life sciences will still be directed to our Opus models.” If your safety documentation describes routing as a cybersecurity mechanism, it’s describing half of it. The same page also restates the safeguard figures for the 5.1 generation and scopes them more narrowly than the August 7 post did. Anthropic says its latest biology safeguards for Fable 5.1 and Fable 5 fire 85% less often for benign requests related to elementary biology and medical questions. It also says Claude Code users can expect an average of around 60% fewer interventions per session from the cyber safeguards, measured against the previous safeguards on Fable 5. Both are vendor figures. One capability change belongs on a security team’s desk rather than a safety team’s. Anthropic says Fable 5.1 can now be used to discover software vulnerabilities, though not to develop exploits for them. On the API, the fallbacks parameter accepts a "default" mode applying Anthropic’s recommended fallback per refusal category, and server-side fallback is in beta. One change here costs money rather than behavior. Pre-output refusals in three categories are billed again from September 24, 2026, and the refusal callout above covers it.

The destination is “our Opus models,” plural and unnamed
Two routed categories: dual-use cybersecurity work and life sciences research and development
Named cyber examples: penetration testing, exploit generation, binary-based vulnerability scanning
Vendor figures: 85% fewer fires on benign elementary biology and medical questions for the 5.1 generation, and around 60% fewer Claude Code interventions per session
Anthropic says Fable 5.1 can discover software vulnerabilities but not develop exploits for them
From September 24, 2026, pre-output refusals in three categories are billed again, on all platforms
Repoint any runbook that names a specific fallback model

Compliance and Admin API mature, and one change reaches backward August 26 to September 24

A records process that was tested in September can be broken today, retroactively, with no error anywhere. On September 24, 2026 the Compliance API Activity Feed stopped returning file names, project document names and artifact titles. The filename and title fields are now always empty or omitted, including on activities recorded before the change. Past records changed shape. Names must now be resolved by ID, using a Compliance Access Key with read:compliance_user_data. If your e-discovery or records workflow reads those fields, re-test it this week. The rest of the arc. On August 26 the Compliance API session endpoints left beta for Claude Cowork and Claude Code. The Admin API arrived in the ant CLI and across all SDKs. The September 1 release notes added the anthropic-version header to the Admin API Claude Enterprise endpoints, the Analytics API and the Compliance API guides. They instruct you to send it on every request. So anything built against the August 26 item may not be sending it. On September 24 local sessions for Microsoft 365 left beta, covering product_surface values beginning office_agents across Excel, PowerPoint, Word and Outlook. Claude in Chrome, at claude_in_chrome, is still beta. Every item here lives on rolling release notes, so confirm each directly before a records process depends on it.

Activity Feed file names, document names and artifact titles are gone, retroactively
Resolve names by ID with a Compliance Access Key carrying read:compliance_user_data
Microsoft 365 local sessions out of beta: Excel, PowerPoint, Word and Outlook
Claude in Chrome transcripts remain beta
Send anthropic-version on every Admin, Analytics and Compliance API request

Cache diagnostics are out of beta September 23

You no longer need a beta header to see what your cache is doing. As of September 23, 2026, cache diagnostics don’t require cache-diagnosis-2026-04-07, and POST /v1/messages responses always include a diagnostics field. Cache reads now run from $1 on Fable 5 down to $0.20 on Opus 5.5 and Sonnet 5.5. So measuring your hit ratio isn’t an engineering nicety any more. It’s the difference between two forecasts. This sits on a rolling release note, so confirm it against your own responses.

No beta header needed; diagnostics ships on every Messages response
Use it to measure the cache-hit ratio before you bank a cache saving
Release-note only: confirm directly

Claude Code enterprise controls Changelog only, and now off-domain

Two models shipped six days apart. If your model allowlist is loose, both of them are already reachable. That’s the setting to look at first. availableModelsMatch now takes the value "exact". That makes an availableModels entry allow only the model version it names. New releases stay blocked until an admin lists them. Alongside it: deniedModels, allowedProviders, and allowClaudeInChromeWithManagedMcp. There’s also a behavior change worth catching before it surprises someone. When an admin removes access to the default model, Bedrock and Vertex sessions now switch to an older available model of the same tier instead of failing. Quiet downgrade rather than loud error. The earlier cluster still stands. Model restriction through managed settings declines an unauthorized switch with an admin-enablement message. maxEffortLevel caps reasoning effort across Bedrock, Vertex and Foundry. That’s a spend control as much as a behavior control. Unparseable managed settings fail startup and name the offending setting. managedMcpServers distributes HTTP and SSE MCP servers organization-wide. And permissions.blockReadsOutsideWorkingDirectories prompts once before the first read outside the working directories. Opus 5.5 is now the default Opus model in Claude Code, and Sonnet 5.5 the default Sonnet model on the Anthropic API.
· Sourcing warning: platform.claude.com/docs/en/release-notes/claude-code now redirects off Anthropic’s own documentation domain to the changelog on GitHub, which is numbered by version and carries no dates at all. No Claude Code claim can be dated from a first-party page any more. Check each item against your own installed version.

availableModelsMatch: "exact" blocks a new model release until an admin lists it
deniedModels, allowedProviders and allowClaudeInChromeWithManagedMcp
Bedrock and Vertex now downgrade to an older same-tier model instead of failing
maxEffortLevel caps reasoning effort; unparseable settings fail startup by name
The Claude Code release notes now redirect off-domain to an undated GitHub changelog

Agent governance Release notes and the ant CLI

An operator can now talk to a running agent instead of watching it finish being wrong. Three additions belong on an agent architect’s desk, and all three sit on rolling pages. Managed Agents permission policies gained an auto mode in which, in Anthropic’s words, “the server evaluates each agent or MCP tool call and runs it, denies it, or pauses for your approval.” A new evaluation field on the agent.tool_use and agent.mcp_tool_use events makes that decision visible in the event stream rather than inferred after the fact. The ant CLI 1.30.0 added ant apply, which declares agents, environments, skills, memory stores and deployments as files with a committed claude-lock.json lockfile. That’s infrastructure as code for an agent estate. It pairs naturally with Skills loading from GitHub: reusable behavior becomes reviewable in source control, and now so does the estate it runs in. The ant CLI 1.32.0 added ant beta:sessions connect, which attaches to a running agent session so an operator can message it and approve or deny in flight. None of these is anchored to a dated article, so confirm all three directly.

auto policies evaluate each tool call and can pause for approval
An evaluation field on agent tool-use events makes the decision auditable
ant apply with a committed claude-lock.json lockfile
ant beta:sessions connect for live attach, message and approve or deny

On-demand compaction September 14 beta

A 1M-token window is capacity, not recall. Compaction is Anthropic’s own admission that a long window is a thing to manage rather than a thing to fill. The Messages API can now return a signed compaction block summarizing the messages you sent, behind the beta compact-2026-09-04. You replay that block in place of the messages it covers. Recent turns are kept verbatim, and on models that preserve thinking, that thinking stays valid. It cuts both ways, though. The cache-hit column is where the money is, and compaction shortens the prompt while changing the prefix. So a workload tuned around a stable cached prefix should measure the net effect before assuming a saving. This is a beta on a rolling release note. Pin it, regression-test it, and confirm it directly.

Signed compaction block, replayable in place of the messages it summarizes
Recent turns retained verbatim; preserved thinking stays valid
Beta header compact-2026-09-04
Compaction changes the prefix, so re-measure your cache-hit ratio

New betas worth knowing API

Three beta surfaces are worth knowing. Per-message effort, behind the header mid-conversation-output-config-2026-07-01, is available on Fable 5.1, Mythos 5.1 and Opus 5. It lets one conversation move up and down the effort ladder instead of committing at the first turn. Turn-scoped system messages narrow a system instruction to a single turn. And thinking.display: "updates" changes how reasoning is surfaced. All three are beta. Regression-test before any of them sits in a critical path.

Per-message effort on Fable 5.1, Mythos 5.1 and Opus 5
Turn-scoped system messages
thinking.display: "updates"
Beta surfaces move; pin and re-test

Claude in Chrome Pro feature and Compliance API surface

A browser agent isn’t a feature of chat. It’s its own environment, and it has a different risk shape from a terminal agent or a governed desktop workspace. The untrusted content is the page itself. The credentials are already in the session. The blast radius is whatever the signed-in browser can reach. Claude in Chrome also sits on both sides of the buying decision. And it’s the surface most likely to be missing from an inventory built before it existed. On the pricing page it’s a Pro plan feature, listed as “Claude in Chrome and Microsoft 365.” And it’s a Compliance API surface. Local session transcript retrieval for Claude in Chrome, with a product_surface value of claude_in_chrome, is still in beta for Claude Enterprise organizations. The Microsoft 365 surfaces left beta on September 24, 2026. That half comes from a rolling page, so confirm it directly before a retention process depends on it.

A Pro plan feature on the pricing page
Compliance API transcript retrieval still in beta for Claude Enterprise, while Microsoft 365 local sessions went GA on September 24, 2026
Page content is untrusted input, and the browser already holds the session
Give it its own row in your environment and boundary inventory

Anthropic’s own pages disagree about zero data retention on 5.1

Two buyers can read Anthropic’s own pages and walk away with different plans. That isn’t a stale page. That’s a live conflict. On one side, the two September 1 news pages say eligible customers will be able to use Fable 5.1 with zero data retention until EFS is ready. Fable 5 too. On the other, the September 1 API release note says both 5.1 models aren’t available under zero data retention unless expressly authorized by Anthropic. The retention documentation says the same thing: not available for any of them unless expressly authorized. Anthropic doesn’t publish the eligibility criteria for the interim offer. It also doesn’t say whether eligible customer and expressly authorized are one status reached two ways or two different things. So read both sides and settle neither on your own, exactly as you should with the Mythos access question. Put it in writing, name both sentences in the email, and get the answer into the contract instead of onto a slide.

Anthropic’s direct terms and Microsoft’s subprocessor terms for Fable 5.1 are not the same

The route you buy through is doing more work here than the model id. The conflict above is Anthropic arguing with itself. This one isn’t. Microsoft treats the two Fable models differently. The extra admin-center controls and the off-by-default gate attach to Fable 5.0. Fable 5.1 can run another way. For certain organizations it’s available with Anthropic acting as a subprocessor, and Microsoft’s Product Terms and Data Protection Addendum (DPA) apply. On that route Microsoft states that Anthropic doesn’t retain customer content, prompts and responses included. Where Microsoft’s page does describe retention, it words the window as up to 30 days. That’s a ceiling. Anthropic’s own documentation says at least 30 days, which is a floor. Inside Microsoft’s services, Microsoft’s Product Terms and DPA are the contract, so Microsoft’s wording governs on the subprocessor route. Buy direct and Anthropic’s governs. The same page now names the data-retaining families as well, “such as Claude Fable 5 and Claude Mythos 5,” so Mythos now sits in Microsoft’s data-retention category too. Now hold that beside Anthropic’s direct terms. There, Fable 5.1 is a Covered Model and 30-day retention is required. The September 1 API release note says the model isn’t available under zero data retention unless expressly authorized by Anthropic. Both pages are accurate about what they describe. That’s the whole problem. Don’t carry Microsoft’s sentence over as Anthropic’s, and don’t assume the subprocessor path is open to you. Settle the prior question first. Which set of terms governs your deployment: direct Anthropic contracting, or a Microsoft commercial agreement with Anthropic as a subprocessor? Answer that, then take the retention answer.
· Source: “Anthropic models in Microsoft Online Services,” Microsoft Learn, a living page stamped September 18, 2026 and read October 1, 2026.

3 / Lineup & Prices

Read the price column, because the names point the wrong way

The newer Opus costs less than the older one. Hold onto that while you read the tables. Opus 5.5 is $4 / $20 per million tokens and Opus 5 is $5 / $25. Fable 5.1 still tops the generally available lineup at $10 / $50, and it’s still retention-required. Sonnet 5.5 and Sonnet 5 both cost $2 / $10, so there’s no price reason to stay on the old one. And the “Legacy models” heading on the pricing page now holds nine models, most of them still Active. So don’t read a label. Read the price, the cache column and the lifecycle row.

Every number in the table below is the standard tier, and there is a second tier

A Claude price quoted without its tier is half a number, right? Every figure in the table is the standard tier on Anthropic’s pricing documentation. The tier matters because Anthropic also runs a batch tier at a 50% discount on both input and output. That’s the tier a large offline workload should be costed against. Fable 5.1 runs at $5 / $25 per million tokens rather than $10 / $50. Opus 5.5 at $2 / $10, Opus 5 at $2.50 / $12.50, Sonnet 5.5 and Sonnet 5 at $1 / $5, Haiku 4.5 at $0.50 / $2.50. Anthropic uses Opus 5.5 as its own worked example: $2 per MTok input and $10 per MTok output against $4 / $20 standard. So if your evaluation compares Claude with another vendor on interactive list price alone, and half your volume runs overnight, you compared the wrong number.

Model	Standard base input	5m cache write	1h cache write	Cache hit	Standard output
Claude Fable 5.1	$10	$12.50	$20	$0.25	$50
Claude Mythos 5.1, invite only	$10	$12.50	$20	$0.25	$50
Claude Opus 5.5	$4	$5	$8	$0.20	$20
Claude Sonnet 5.5	$2	$2.50	Unconfirmed	$0.20	$10
Claude Fable 5	$10	$12.50	$20	$1	$50
Claude Mythos 5, invite only	$10	$12.50	$20	$1	$50
Claude Opus 5	$5	$6.25	$10	$0.50	$25
Claude Opus 4.8	$5	$6.25	$10	$0.50	$25
Claude Sonnet 5	$2	$2.50	$4	$0.20	$10
Claude Haiku 4.5	$1	$1.25	$2	$0.10	$5

The cheapest cache read in the lineup is no longer on the most expensive model

Read the cache-hit column, not the headline. It reorders the lineup. Fable 5.1 charges $10 per million input tokens, exactly what Fable 5 charges, and reads cached context at $0.25 against Fable 5’s $1. That alone was the whole Fable 5.1 cost story a month ago. It isn’t now. Opus 5.5 and Sonnet 5.5 both read cached context at $0.20, below Fable 5.1, while Opus 5 charges $0.50 and Fable 5 charges $1. So a cache-heavy agent that used to pick Fable 5.1 for the read price should re-run the math. Opus 5.5 is cheaper on base input, cheaper on output and cheaper on the read. Most agent work re-reads a large stable context on every call, and so do most document assistants and most long-running coding sessions. Model your own cache-hit ratio before you bank any of this, because a workload that never reuses a prefix sees none of it. One cell in the table isn’t settled. The one-hour cache write for Sonnet 5.5 wasn’t published in the row read on October 1, 2026, so confirm it on the pricing page before it enters a model.

Model	Where it sits	Status on October 1, 2026
Claude Fable 5.1	Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud and Microsoft Foundry	Active, floor of “Not sooner than September 1, 2027”; the only Fable model in the models overview comparison table
Claude Opus 5.5	Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, Microsoft Foundry; Pro, Max, Team and seat-based Enterprise	Released September 22, 2026; Active, floor of “Not sooner than September 22, 2027”; not a Covered Model, so no retention requirement
Claude Sonnet 5.5	Current Sonnet; named as the migration target for Sonnet 4.5	Released September 28, 2026; Active, floor of “Not sooner than September 28, 2027”; not a Covered Model
Claude Mythos 5.1	Not generally available; access through Project Glasswing or an Anthropic, AWS or Google Cloud account team	Active (invite only), floor of “Not sooner than September 1, 2027.” The model page states it is the same model as Claude Fable 5.1, offered separately
Claude Mythos 5	Not generally available	Active, floor of “Not sooner than June 9, 2027”
Claude Fable 5	Still selectable and still priced at $10 / $50	Under the “Legacy models” heading on the pricing page, but Active on the deprecations page, not sooner than June 9, 2027
Claude Opus 5	Still selectable and still priced at $5 / $25, above the newer Opus 5.5	Released July 24, 2026; now under “Legacy models” on the pricing page, Active on the deprecations page, not sooner than July 24, 2027
Claude Opus 4.8	Still selectable and still priced, at exactly the Opus 5 rate	Legacy on the pricing page, Active on the deprecations page, not sooner than May 28, 2027
Claude Sonnet 5	Still selectable at $2 / $10, the same price as Sonnet 5.5	Now under “Legacy models” on the pricing page, Active on the deprecations page, not sooner than June 30, 2027
Claude Sonnet 4.5	Serving until the retirement date, then not at all	Deprecated September 30, 2026, with retirement on the Claude API scheduled for November 30, 2026. Anthropic recommends migrating to Claude Sonnet 5.5
Claude Haiku 4.5	Not a plan default; the cheapest priced tier	Active, but `claude-haiku-4-5-20251001` carries a tentative retirement floor of “Not sooner than October 15, 2026,” 14 days out
Claude Opus 4.1	Nowhere; the identifier stops serving requests	Retired August 5, 2026; Anthropic’s current migration pointer sends Claude Opus 5 and earlier to Claude Opus 5.5

Fable 5.1 and Mythos 5.1 launched on September 1, 2026

Thinking is no longer something you switch on for the hard tasks. It’s always on. The model ids are claude-fable-5-1 and claude-mythos-5-1, and both carry 1M token context by default, 128k maximum output and always-on adaptive thinking. Anthropic describes them as “the world’s most advanced models for coding and knowledge work.” That’s a vendor claim. Test it on your own tasks before it enters a business case. Availability is broad from day one: the Claude API, Amazon Bedrock, the Claude Platform on AWS, Google Cloud and Microsoft Foundry. Mythos 5.1 is invite only, and Anthropic’s own model page says it’s the same model as Fable 5.1 offered separately, so don’t plan around it. Fable 5.1 is still the top of the generally available lineup at $10 / $50. Work from the pricing table above, because two cheaper models have landed under it since.

Claude Fable 5.1 Still the top tier

The defaults are the story here. 1M token context is the default now, not a beta flag. Maximum output is 128k. And adaptive thinking is always on, so there’s no configuration in which this model isn’t reasoning before it answers. It’s still the top of the generally available lineup, at model id claude-fable-5-1. Anthropic’s models overview now reserves it for demanding reasoning and long-horizon agentic work, or for when your evals on Opus 5.5 at higher effort still fall short. Anthropic positions Fable 5.1 and Mythos 5.1 together as “the world’s most advanced models for coding and knowledge work.” That’s a vendor claim. Treat it as one until your own evaluation says otherwise.

1M token context by default; 128k maximum output
Always-on adaptive thinking, not a toggle
Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, Microsoft Foundry
$10 / $50 per million tokens, with cache hits at $0.25
A Covered Model: 30-day retention required, with a per-workspace override for ZDR organizations

Claude Opus 5.5 New, September 22

The capable tier for work that can’t sit under a retention window. Opus 5.5 is claude-opus-5-5, at $4 / $20 per million tokens with a $0.20 cache hit, 1M token context and 128k maximum output. That rises to 300k on the Batch API in beta. Thinking is always on and adaptive, and the default effort is medium. Fast mode is a research preview on the Claude API. Anthropic says it performs at Fable 5.1 level on most work, costs 40 percent less to run than Opus 5 and produces output more than 30 percent faster. Vendor claims, every one. What isn’t a vendor claim is the list of Covered Models, and Opus 5.5 isn’t on it. Anthropic says zero data retention is available. Three breaking changes against Opus 5. Thinking can’t be turned off at all now: sending disabled fails, and so does setting your own budget_tokens. tool_choice types any and tool return a 400. And computer_20251124 returns a 400 on the Claude API and Google Cloud, where you need computer_toolset_20260801 instead, though Anthropic says the old tool keeps working on Amazon Bedrock.

$4 / $20 per million tokens; cache hit $0.20, below Fable 5.1’s $0.25
Not a Covered Model: no 30-day retention requirement, and ZDR available
1M token context, 128k maximum output, 300k on Batch API beta
Thinking cannot be disabled and a manual budget_tokens also returns a 400
On Pro, Max, Team and seat-based Enterprise with increased five-hour usage limits

Claude Sonnet 5.5 New, September 28

Same price as Sonnet 5, different behavior, and the named landing place for everyone leaving Sonnet 4.5. Sonnet 5.5 is claude-sonnet-5-5, at $2 / $10 per million tokens with a $0.20 cache hit, 1M token context and 128k maximum output. Anthropic says it’s more than 30 percent faster than Sonnet 5 and costs up to 30 percent less for most work. Vendor claims. The default effort differs by surface: medium in the Claude apps, high on the Claude Platform. So the same prompt is a different bill depending on where it runs, and nothing in your code shows that. Five changes break Sonnet 5 code. Forced tool use returns a 400. Disabling thinking at high effort or below is now thinking: {"type": "between_tools"} rather than "disabled". And the advisor tool rejects Claude Opus 4.8, Claude Opus 4.7 and Claude Sonnet 5 as advisors, which is an agent-architecture change dressed as a model release. Sonnet 5.5 isn’t a Covered Model, and Anthropic says zero data retention is available on all platforms.

$2 / $10 per million tokens; cache hit $0.20
Default effort is medium in the Claude apps and high on the Claude Platform
The advisor tool rejects Opus 4.8, Opus 4.7 and Sonnet 5 as advisors
Forced tool use returns a 400, same as on the 5.1 pair and Opus 5.5
The migration target Anthropic names for Claude Sonnet 4.5

Claude Mythos 5.1 Invite only, with a floor

Three first-party sources describe how you get access to Mythos 5.1, and none of them beats the other two. It’s the gated tier, at model id claude-mythos-5-1, sharing Fable 5.1’s specification and price. The model page now settles the part that mattered most: Mythos 5.1 is the same model as Claude Fable 5.1, offered separately through Project Glasswing. Its status is Active (invite only), and the page tells you to contact your Anthropic, AWS or Google Cloud account team. The deprecations table now carries it too, with a floor of not sooner than September 1, 2027, and Mythos 5 at not sooner than June 9, 2027. So the lifecycle question is answered. The access question isn’t. The September 1 announcement names two programs. The Life Sciences Verification Program is designed so that life sciences professionals can use Claude Mythos 5.1 with safeguards built for professional research and development activities. The Cyber Verification Program is described differently. It “currently provides access to certain Opus- and Sonnet-class models with reduced cyber safeguards for defensive security work.” Read that literally. It gates Opus- and Sonnet-class models, not Mythos 5.1. Then there’s a third framing: the September 1 API release note still describes claude-mythos-5-1 as being for Project Glasswing participants. On September 17, 2026 the LSVP got its own article with a grant structure attached. Verification includes a review of research credentials, security standards and ethical research oversight. Two grant types: Standard Use, renewed annually, and High-risk Use, renewed every six months. Standard Use grants apply to Mythos 5.1, Opus 5 and Sonnet 5 today, and to future models as they launch. Note what that list doesn’t contain. The LSVP page hasn’t been updated for Opus 5.5 or Sonnet 5.5, so don’t assume a grant reaches them. A September 30, 2026 changelog entry on the same page adds one useful clarification: the form it links to is an expression of interest, not an application. Which is the clearest argument yet for asking your account team by name.

The model page states Mythos 5.1 is the same model as Claude Fable 5.1, offered separately
Status Active (invite only), with a floor of not sooner than September 1, 2027
Three first-party framings: the Cyber Verification Program, the Life Sciences Verification Program and Project Glasswing
The CVP as described gates Opus- and Sonnet-class models, not Mythos by name
LSVP Standard Use grants name Mythos 5.1, Opus 5 and Sonnet 5, not the 5.5 models
Not generally available, so it is not a procurement option today

Claude Opus 5 Outclassed by its own successor

If a routing rule in your estate sends regulated traffic to Opus 5, it’s now sending it to the slower, dearer option. Opus 5 was released July 24, 2026 at $5 / $25 per million tokens with a $0.50 cache hit, 1M context and 128k maximum output. For two months it was the most capable Claude model with no mandatory retention requirement. Opus 5.5 took that position on September 22 at $4 / $20 with a $0.20 cache hit, with Anthropic claiming Fable 5.1 level performance and zero data retention availability. Cheaper on input, cheaper on output, far cheaper on cache reads. Opus 5 has also moved under the “Legacy models” heading on the pricing page while staying Active on the deprecations page with a floor of not sooner than July 24, 2027. So it’s still selectable, still priced, still supported. It’s just no longer the answer to the question it used to answer. One comparison a buyer should still see: on the standard table, Opus 5 and Opus 4.8 are priced identically, line for line, and both now cost more than Opus 5.5.

$5 / $25 per million tokens on the standard tier; cache hit $0.50
No data-retention requirement, but Opus 5.5 offers the same at a lower price
Now under “Legacy models” on the pricing page, Active to not sooner than July 24, 2027
Priced identically to Opus 4.8 across every column
Re-point regulated routing rules at Opus 5.5 and measure the difference

Claude Fable 5 Legacy in marketing, Active in lifecycle

Legacy in marketing. Active in lifecycle. Still priced and still selectable. A procurement policy that keys off the word “legacy” will read the marketing page and stop, so give a buyer all three parts. Fable 5 hasn’t been retired and hasn’t been repriced: $10 / $50 per million tokens on the standard tier, with a $1 cache hit. What changed is who is editorializing. The pricing page used to read as the unopinionated inventory beside an opinionated models overview, and it’s neither of those now. It carries a heading reading “Legacy models,” and that heading has been getting crowded. It now opens with Sonnet 5 and Opus 5, then Fable 5, then Opus 4.8, Sonnet 4.6, Opus 4.7, Opus 4.6, Sonnet 4.5 and Opus 4.5. Most of those are still Active on the deprecations page, Fable 5 among them, with a tentative retirement of not sooner than June 9, 2027. One isn’t: Sonnet 4.5 is deprecated and retires November 30, 2026. So the heading tells you nothing about whether a model is about to stop serving. Neither page is wrong. They answer different questions. Fable 5 is also one of the four Covered Models, designated on June 9, 2026, so it carries the same 30-day retention requirement as the 5.1 pair and the same per-workspace override. Both pages are living documents, so check them on the day.

Still active and still priced at $10 / $50 on the standard tier
Under the “Legacy models” heading on the pricing page, which now holds nine models
Active on the deprecations page, not sooner than June 9, 2027
Cache hits cost $1, four times the 5.1 rate and five times the Opus 5.5 rate
A Covered Model since June 9, 2026, so retention-required

Claude Sonnet 5 Same price, newer sibling

Sonnet 5 costs exactly what Sonnet 5.5 costs. That’s the whole decision. $2 / $10 per million tokens with a $0.20 cache hit on both, so there’s no price reason to stay. The pricing question that hung over Sonnet 5 was settled by the calendar rather than by a promise: the scheduled September 1 increase didn’t happen, and $2 / $10 is the standard price. Since then Sonnet 5 has moved under “Legacy models” on the pricing page while staying Active on the deprecations page to not sooner than June 30, 2027. There’s one reason to look before you jump. Moving to Sonnet 5.5 isn’t a drop-in: five changes break Sonnet 5 code, including forced tool use and the advisor tool refusing Sonnet 5 itself as an advisor. Budget the migration, not just the tokens.

$2 / $10 standard, the same as Sonnet 5.5; $1 / $5 on the batch tier
Now under “Legacy models” on the pricing page, Active to not sooner than June 30, 2027
Five breaking changes stand between Sonnet 5 code and Sonnet 5.5
The advisor tool on Sonnet 5.5 rejects Sonnet 5 as an advisor
No current page restates a per-plan default at version granularity

Claude Opus 4.8 Two pages, two labels

The legacy label isn’t on the page a buyer would check for it. The models overview no longer has a Legacy heading at all. Its four-row comparison table is now Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 4.5. And the migration pointer on it has moved twice in ten weeks. It now tells anyone on Claude Opus 5 or earlier to read Anthropic’s migration notes for Claude Opus 5.5. That sentence quietly includes Opus 4.8. The same page tells an undecided buyer to start with Claude Opus 5.5 for most workloads. Fable 5.1 is for demanding reasoning and long-horizon agentic work, or for when evals on Opus 5.5 at higher effort still fall short. A different first-party page carries the label. Under “Legacy models” on the pricing page, Opus 4.8 sits at $5 / $25. Beside it are Sonnet 4.6 and Sonnet 4.5 at $3 / $15, and Opus 4.7, 4.6 and 4.5 at $5 / $25. And the deprecations page still shows Opus 4.8 as Active, not sooner than May 28, 2027. So the honest three-way position is legacy in marketing, active in lifecycle, still priced. And priced identically to Opus 5 on the standard table, which means both now cost more than the newer Opus 5.5. Opus 4.8 isn’t described as a safety fallback anywhere readable. One more thing an agent architect should catch: on Sonnet 5.5 the advisor tool rejects Opus 4.8 as an advisor outright.

Legacy on the pricing page, Active on the deprecations page, still priced
Identical standard pricing to Opus 5, and both cost more than Opus 5.5
The migration pointer now sends Claude Opus 5 and earlier to Claude Opus 5.5
The models overview recommends starting with Opus 5.5 for most workloads
On Sonnet 5.5, the advisor tool rejects Opus 4.8 as an advisor

Read the price column, not the model name and not the version number

The names are lying to you, and now the version numbers are too. A buyer reading the lineup cold assumes Fable is the mid-tier and Opus is the premium one, because that’s how these names have worked everywhere else. It’s the other way round. Fable 5.1 costs $10 / $50 per million tokens. Opus 5.5 costs $4 / $20. Opus 5 costs $5 / $25. Sonnet 5.5 and Sonnet 5 both cost $2 / $10. Haiku 4.5 costs $1 / $5. Read that list again. The newer Opus is cheaper than the older Opus, by 20 percent, while Anthropic claims it’s the better model. A higher version number went down in price, which is the opposite of how every other line in your software budget behaves. Now add the cache column and a third ordering appears. Opus 5.5 and Sonnet 5.5 read cached context at $0.20, below Fable 5.1 at $0.25, well below Opus 5 at $0.50 and five times below Fable 5 at $1. So for a cache-heavy workload the gap between the tiers is narrower than the headline suggests, and in places it points the wrong way. One more inversion sits inside the lineup. Opus 4.8, which the pricing page files under “Legacy models,” costs exactly what Opus 5 costs: $5 / $25 with a $0.50 cache hit. Both are dearer than the current Opus. The word “legacy” buys you nothing here, and neither does the word “new.” Every figure in this paragraph is the standard tier. The batch tier halves both input and output. Assign tiers by measuring your own traffic mix, not by reading the names.

Sonnet 5 is still $2 / $10, and the scheduled rise did not happen

If you rebuilt a Sonnet 5 budget at $3 / $15, tear it up. Anthropic canceled the increase that had been scheduled for September 1, 2026. September 1 came and went and the price didn’t move. $2 / $10 per million tokens is the standard price now, not introductory pricing, and you can stop hedging. Two qualifiers survive the good news. A price that has stopped being introductory is an ordinary list price with no committed floor, and that’s a weaker guarantee than a contract. And Sonnet 5 has since moved under the “Legacy models” heading on the pricing page, alongside Opus 5. It stays Active on the deprecations page, with a floor of not sooner than June 30, 2027. Sonnet 5.5 is the current Sonnet, at the same $2 / $10.
· Anthropic’s wording: “The $2/$10 per million input/output token pricing for Claude Sonnet 5, announced at launch as introductory pricing through August 31, 2026, is now the standard price. The previously scheduled increase to $3/$15 per million input/output tokens on September 1, 2026 will not occur.”

Plan	Price as published	Model availability	Notable inclusions
Free	No cost	Sonnet and Haiku. Fable is shown as “No”	Family names only; the page never names a version number
Pro	The page gives two annual figures that do not reconcile, $17 per month billed annually and $200 up front, against $20 monthly	Opus, Sonnet and Haiku, with Fable on “Usage credits.” Opus 5.5 is included, with increased five-hour usage limits	Claude Code; Claude Design, Slides and Docs; Claude in Chrome and Microsoft 365. The Claude Code default model on Pro moved from Sonnet to Opus
Max	From $100 per month	Choose 5x or 20x more usage than Pro; Opus 5.5 included with increased five-hour usage limits	Higher output limits for all tasks and priority access at high traffic times
Team	$20 per seat per month billed annually, $25 monthly; premium seats $100 billed annually, $125 monthly	Premium seats carry 5x more usage than standard seats; Opus 5.5 included	For teams of 2 to 150; includes Claude Code and Claude Cowork. The Claude Code default on Team Standard moved from Sonnet to Opus
Enterprise	Seat price plus usage at API rates. US$20 per seat per month billed annually, as read on claude.com/pricing on October 1, 2026; identical to a standard Team seat and unconfirmed by Anthropic	Not itemized by family on the pricing page; Opus 5.5 is available on seat-based Enterprise	The Compliance API surfaces. Microsoft 365 local sessions left beta on September 24, 2026; Claude in Chrome transcripts are still beta

Which pricing page to read, and what it will not tell you

A page that refuses your request isn’t telling you the information is unpublished. Anthropic puts pricing on two hostnames and only one of them answers. anthropic.com/pricing returns a 406 and can’t be fetched at all. claude.com/pricing renders, and it carries the plan and seat structure set out above. So read the one that works. Then keep pinning a model id. That page names model families, Opus, Sonnet, Haiku and Fable, and never version numbers. And the only places it uses the word “default” are about training rather than model selection. “Sonnet 5 is the Free default” and “Opus 5 is the Max default” are still July 24 claims that no current page restates at version granularity. One default did move, and it moves money. Claude Code changed the default model on Pro and Team Standard plans from Sonnet to Opus, matching Max, Team Premium and Enterprise. Anyone who budgeted Pro or Team Standard Claude Code usage against a Sonnet default is now paying for an Opus default. One more consequence deserves its own line. Fable on Pro is described elsewhere as 50% of weekly limits. That’s a usage allowance, not an entitlement, and it’s still enough that no Opus model is the strongest thing a Pro subscriber can reach. Two cells in the plan table are transcriptions rather than settled prices, and they’re marked as such there. The Pro row carries both $17 per month billed annually and $200 up front. Twelve months at $17 is $204. So one of those two figures is rounded, stale or describing something the page never spells out. Both are printed here as the page gives them. The Enterprise row quotes US$20 per seat per month billed annually. That’s the same number as a standard Team seat, and one fifth of a Team premium seat. It sits on a row whose model availability cell says the page doesn’t itemize Enterprise at all. Neither number is safe in a business case until Anthropic confirms it.
· Both cells were read on claude.com/pricing on October 1, 2026. The pricing page is a living document, so check it on the day.

Claude for Teachers Dated, with a changelog

Verified US K-12 educators get free premium access, and the sign-up window runs through June 30, 2027. This one is easier to trust than most. The article is dated July 14, 2026 and carries a published changelog, with updates stamped July 21 and August 28, 2026. So a change to the window would leave a trace on the page rather than happening silently.

Announced July 14, 2026, with dated updates July 21 and August 28, 2026
Sign up by June 30, 2027 for a full year of access
Verification required
A recruiting channel as much as a product

Self-serve HIPAA Confirm the scope

Self-serve is a process change, not a scope change. HIPAA configuration became self-serve for Enterprise and Platform customers on July 14, 2026, which takes a manual step out of healthcare deployments. Read it beside the Covered Model retention requirement, the per-workspace override and the platform split on where retained data sits, before you assume the two compose.

Configure without a bespoke process
Still requires your own compliance review
Read it beside the Covered Model retention requirement
Confirm scope before moving regulated data

Three facts procurement will ask about

Start with retention, because it’s the fact most often stated wrongly. Fable 5.1, Mythos 5.1, Fable 5 and Mythos 5 are the four Covered Models, and all four require 30-day data retention. Anthropic says the data isn’t used for training and is retained for at least 30 days before automatic deletion, with carve-outs for a safety investigation and for legal requirements. The retention itself is mandatory. It isn’t an all-or-nothing gate on a zero data retention organization. Anthropic documents enabling 30-day retention for a single Console workspace while every other workspace keeps zero data retention. Everything outside those four models carries no requirement at all, and that contrast is what decides tiering for regulated work. The contrast just got much easier to live with. Opus 5.5 isn’t a Covered Model, and Anthropic says zero data retention is available on it. Anthropic also claims it performs at Fable 5.1 level on most work for $4 / $20 against Fable 5.1’s $10 / $50. So regulated traffic that can’t sit under a 30-day window no longer has to accept a big capability drop to stay outside the retention gate. It accepts a vendor claim instead, which you should test. Anthropic has separately published an interim zero data retention offer for eligible customers, and the EFS program as the intended permanent answer. It hasn’t published the eligibility criteria for either. Second, continuity. Access to Fable 5 and Mythos 5 was suspended on June 12, 2026, the redeployment notice was published June 30, and redeployment began July 1. So build your plan on the assumption that a top tier can be withdrawn. Third, defaults, and they move. The July 24 announcement made Opus 5 the Claude Max default and Sonnet 5 the Free and Pro default. No current page restates a default at version granularity. Meanwhile Claude Code changed its default model on Pro and Team Standard from Sonnet to Opus, and made Opus 5.5 the default Opus. A plan default isn’t a stable thing to budget against. Pin a model id. And read the hostname that answers: anthropic.com/pricing still returns a 406 and can’t be fetched, while claude.com/pricing renders.

4 / Product Family

Claude is now a family of work surfaces

Choosing Claude isn’t choosing a model any more, right? The real question is where the work should run. Conversation, terminal, governed desktop workspace, team channel, design and document surface, the browser, or a business application. The browser is the newest of those, and the one most likely to be missing from an inventory you already built.

Analyze and write

Claude chat with Sonnet 5.5, Opus 5.5 or Fable 5.1

Use the main Claude experience for source-grounded analysis, research, writing and ready-to-use files. The pricing page puts Sonnet and Haiku on Free and shows Fable there as “No.” Pro adds Opus and reaches Fable on usage credits. Max buys 5x or 20x more usage than Pro. Fable 5.1 is the most capable tier, and it’s a Covered Model, so 30-day retention comes with it. Opus 5.5 isn’t a Covered Model, so it carries no retention requirement at all.

Attach the complete source pack
Choose effort by consequence
Check the workspace retention setting before you pick 5.1
Pick Opus 5.5 when retention is the constraint

Surface	Best for	Context advantage	Primary caution
Claude chat	Analysis, writing, research and files	Conversation, uploads, connectors and memory	Large context still needs coverage checks
Claude Code	Repository-scale engineering	Codebase, terminal, tools and project instructions	Command and prompt-injection risk
Claude Cowork	Desktop knowledge work	Selected folders, local apps, skills and plugins	Mount permissions and action review
Claude Tag	Shared team delegation	Channel history, scoped tools and shared memory	Spend, permissions and public mistakes
Claude in Chrome	Work that lives in a web application	The open page, the signed-in session and the browser’s own reach	Page content is untrusted input; transcript retrieval is beta
Claude Design	Rapid visual exploration	Brand inputs, conversation and prototypes	Still a Labs product; professional finishing required
Claude Slides and Claude Docs	Decks and written deliverables	The analysis already in the conversation	A bundled Pro feature, not a separately governed surface
Claude Science	Auditable scientific work	Packages, tools, compute and artifacts	Expert validation remains essential

5 / Inside Work

From source material to finished work without changing context

People rebuild the same brief four times a day in four different tools. That’s the waste Anthropic is aiming at. Claude is meant to follow the work across files, applications and teams instead of making anyone start over on each surface.

Engineering

Claude Code

Plans, edits, tests and debugs inside the real repository.

Agent teams and subagents
Long-running workflows
Code review and verification

Knowledge work

Claude Cowork

Runs multi-step tasks against selected desktop files and applications.

Read-only or controlled write modes
Reusable skills and plugins
Dispatch from text or voice

Microsoft 365

Excel, PowerPoint and Word

Claude works directly in business documents and carries context between applications. Write tools shipped on July 7, 2026: Claude can draft and send email, act on the calendar and create or update files.

Model, audit and explain spreadsheets
Build decks from live analysis
Review write permissions before granting them

Collaboration

Claude Tag

A shared team agent in Slack with scoped identity and memory.

Channel-specific access
Shared continuation
Logs and budget controls

Browser

Claude in Chrome

Works inside the browser on web applications that never exposed an API. A Pro plan feature, and now a Compliance API surface in beta for Claude Enterprise.

Read and act on the open page
Page content is untrusted input
Scope the signed-in session deliberately

Creative

Claude Design, Slides and Docs

Explores prototypes and polished visual directions through conversation, and produces decks and documents in the same place. The pricing page bundles the three together as a Pro feature.

Software experience ideation
Slides, docs and one-pagers
Canva handoff

Connectors

MCP and apps

Pulls live context from enterprise and creative systems.

Google Drive and Gmail
Market and portfolio data
Blender and specialist tools

Reusable behavior

Skills and plugins

Packages instructions, tools and workflows into repeatable organizational capability. The Skills API left beta across the SDKs on August 27, 2026.

Reduce prompt folklore
Standardize recurring work
Deploy domain templates

Continuity

Memory and context

Carries relevant context across sessions and, increasingly, across applications. On July 10, 2026 memory was reworked into individual categorized entries, replacing the daily memory summary.

Preserve audience and standards
Inspect and remove entries one at a time
Review what should persist

6 / Creative Work

Claude is becoming useful before, during and after the creative tool

Claude doesn’t replace taste. What it does is widen the number of directions a creative team can try, take the repetitive production off their hands, and keep the strategy attached to the thing being made.

Claude Design Labs

Explore software experiences, prototypes, slides and one-pagers with visual iteration inside the conversation.

Start with audience and desired behavior
Include brand rules and references
Export into professional finishing tools

Creative connectors MCP

Claude can work with tools such as Blender through connectors, scripts and the application’s own APIs.

Analyze and debug scenes
Batch repetitive changes
Create custom tools inside the creative application

Cross-format production

Claude can turn research into a document, document into deck, data into charts and creative brief into a buildable prototype.

Keep one narrative thread
Reduce manual reformatting
Use humans for taste and final selection

Prototype brief

Build three distinct directions for this onboarding flow. Use our brand system, explain the behavioral idea behind each route, and make every prototype testable rather than decorative.

A strong brief names the decision the prototype is supposed to help the team make.

Production automation

Inspect this Blender scene and identify the repetitive production steps. Propose a reversible automation plan, then implement only the steps I approve.

Keep approval between analysis and modification.

Deck from evidence

Turn this research and workbook into a 12-slide executive deck. Use the approved master, keep one claim per slide, and link every chart to its source.

Ask Claude to maintain provenance, not merely visual polish.

Taste-preserving review

Do not redesign this. Review it as a creative director: identify where the execution departs from the stated concept, audience or brand behavior.

Claude is often more useful as a critic than as an automatic stylist.

7 / Behavior Playbook

Eight habits that make Claude substantially better

A better model doesn’t make a team better, right? Behavior does that. The upgrade matters, but what people do with it decides whether all that capability turns into useful work, a bigger bill or a real risk.

Front-load the context

Give Claude the complete source pack and explain how the files relate before asking for conclusions.

Let it inspect the brief

Ask what’s missing or contradictory before it starts producing anything.

Choose effort deliberately

Use lower effort for reversible work and escalate only when consequence justifies it. On 5.1, adaptive thinking is always on, so effort is the only lever left.

Define completion

Name the artifact, the checks it has to pass and the evidence report. Not just the topic.

Keep context across stages

Move from analysis to file to deck without rebuilding the narrative in a new chat.

Use an adversarial second pass

Ask Claude to name the weakest claims, the broken assumptions and the missing tests in its own work.

Name the audience and stakes

Claude changes depth, tone and caution when it knows who will use the result and what failure costs.

Trust the loop, inspect the result

Don’t micromanage every step. Do review the commands, diffs, formulas and finished artifacts.

Outcome delegation

Produce the completed board memo, not an outline. Validate every recommendation against the attached evidence and include a one-page appendix of unresolved assumptions.

Agentic coding

Add pagination to the users API, update the contract and documentation, add edge-case tests, run the full suite, and stop only when CI passes or you can name the blocker precisely.

Self-review

Re-read your work as a skeptical domain expert. Identify the three conclusions most likely to be wrong and show what evidence would change each one.

Long-context coverage

Before synthesizing, create a coverage map of every supplied document. After the answer, list any sources or sections you did not materially use.

8 / Watch Outs

The more Claude can act, the more its environment matters

A weak chat answer is annoying. A tool-using agent with broad file and network access can do real operational damage. Different failure mode, right? Two more categories belong here. One, a model generation that fails requests your current code makes legally today. Two, a set of first-party contradictions a control document can inherit without anyone noticing.

Scope your breaking-change sweep to model ids, not to “the 5.1 models”

A sweep that was correct in September is wrong today, and that’s the point. tool_choice types any and tool now return a 400 on four models, not two: Fable 5.1, Mythos 5.1, Opus 5.5 and Sonnet 5.5. Opus 5.5 adds two more. Thinking can’t be turned off at all, so both disabled and a manual budget_tokens fail. And computer_20251124 returns a 400 on the Claude API and Google Cloud, where computer_toolset_20260801 is required, while Anthropic says the old tool keeps working on Amazon Bedrock. Sonnet 5.5 adds five of its own, including thinking: {"type": "between_tools"} in place of "disabled" at high effort or below, and an advisor tool that refuses Opus 4.8, Opus 4.7 and Sonnet 5 as advisors. Separately, the four Covered Models require 30-day retention, so a request from an organization or workspace without retention enabled returns a 400 invalid_request_error. That last one is a configuration problem with a documented fix, not a wall: see the workspace override above. Test every path before you switch a model id in production.

Zero data retention is not a hard gate on 5.1

The 30-day requirement on Fable 5.1 and Mythos 5.1 reads like a wall. It isn’t one. A zero data retention organization can turn on 30-day retention for a single Console workspace while every other workspace stays at zero. The path is Settings, then Workspaces, then that workspace’s Privacy controls tab. If a procurement file anywhere says no configuration flag fixes this, that’s the sentence to go and fix first. The requirement is real and the failure mode is a clean 400. So the work is a configuration decision and a scoping decision, not an exception request.

Nobody publishes who is eligible for interim zero data retention

Anthropic’s September 1 news pages say eligible customers will be able to use Fable 5.1, and Fable 5, with zero data retention until Enterprise Frontier Safeguards is ready. The September 1 API release note and the retention documentation both say the models aren’t available under zero data retention unless expressly authorized by Anthropic. No page defines eligibility. No page says whether eligible and expressly authorized are the same status. This is an unresolved conflict between first-party sources rather than a stale claim, and it’s published here as one. Don’t let either sentence into a control document on its own.

The retention answer changes with the route you buy through

Read this beside the two cards above, because it’s a different kind of conflict. Those are Anthropic pages disagreeing with each other. This is Anthropic’s direct terms and Microsoft’s subprocessor terms describing the same model and not reaching the same outcome. On Anthropic’s own pages, Fable 5.1 is a Covered Model with a 30-day retention requirement. On Microsoft’s side the off-by-default retention gate attaches to Fable 5.0. And for certain organizations Fable 5.1 runs with Anthropic as a subprocessor under Microsoft’s Product Terms and Data Protection Addendum. There, Microsoft states that Anthropic doesn’t retain customer content. Neither source resolves the other. Don’t move Microsoft’s sentence into an Anthropic control description, or the reverse. And don’t let a procurement file inherit a retention posture from whichever page it read first. Establish the contracting route, then take the retention answer. That same Microsoft page is the only place the seven-year classification-score retention appears, which is a second reason to read it rather than summarize it.
· Source: “Anthropic models in Microsoft Online Services,” Microsoft Learn, a living page stamped September 18, 2026 and read October 1, 2026.

EFS is a promise with no dated milestone, and the two pages disagree on tense

A plan with no delivery date is a different animal from a plan with a date that slips. Enterprise Frontier Safeguards is described as giving privacy equivalent to zero data retention, by keeping customer data on the customer’s own cloud infrastructure. The scope named is Claude Code, Claude Enterprise, the Claude Platform, Amazon Bedrock, the Claude Platform on AWS, the Gemini Enterprise Agent Platform and Microsoft Foundry. The September 1 article says the rollout happens in phases starting later this fall. The request form says EFS is rolling out to eligible customers in phases. Future tense on one page, present tense on the other, and no phase, customer or date on either. Nothing published between September 17 and October 1 announced a phase. Whether any phase has shipped is an open question, and it’s open because the only two pages that would carry the answer contradict each other.

Prompt injection

Instructions inside email, documents, websites or pasted prompts can redirect an agent. Treat external content as untrusted input.

Mount and write access

Cowork permissions should match the task. Read-only, read-write and no-delete boundaries are meaningful controls.

Higher effort cost

More reasoning helps on hard work. It also adds latency, token use and the odd pointless detour. On 5.1 and on Opus 5.5 you can’t turn thinking off at all. And on Sonnet 5.5 the default effort is medium in the Claude apps but high on the Claude Platform. So the same prompt costs more through the API than it does in the app.

Preview instability

Dynamic workflows, Labs products and beta surfaces require regression testing before production use. Per-message effort, turn-scoped system messages and the updates thinking display are all beta.

Long context omissions

A 1M-token window is capacity, not proof of complete recall or equal attention. That’s now the default window on 5.1 rather than an opt-in, so the coverage discipline applies to every call.

No native media generation

Claude can plan, analyze and prototype visuals, but final image, video and audio generation still depends on other tools.

Model-version drift, and one failure that returns a 200

A model upgrade can change formatting, tool choices and reasoning style. Re-test critical prompts and workflows. Watch for the quiet one. A thinking block that Sonnet 5.5 produced works only in the account that produced it or a linked account. A block sent from any other account is dropped before the model sees it, and the request still succeeds. No error, no warning, just a slightly worse answer. Anywhere you share cached assistant turns across accounts, compare outputs rather than status codes.

Diplomatic language

Claude may soften criticism unless directness is explicitly requested. Ask for the harshest defensible review.

Opus 4.1 is gone, and the replacement chain has moved twice

If you haven’t hit this error yet, your pinned job isn’t running, and that’s its own finding. On Anthropic’s deprecations page, read October 1, 2026, claude-opus-4-1-20250805 still shows status Retired with the line “This model was retired August 5, 2026.” Anything still pinned to the 4.1 identifier, an integration, a saved evaluation or a scheduled job, is failing now as a request error rather than a warning. Watch the destination, because it keeps moving. The deprecations row still names claude-opus-4-8 as the replacement. The models overview now says anyone on Claude Opus 5 or earlier should migrate to Claude Opus 5.5. That’s a different target from the one it named in July. And Opus 4.8 and Opus 5 are both filed under “Legacy models” on the pricing page. Migrating 4.1 to 4.8 today buys you one hop into a bucket you’ll have to leave twice.

A floor is not an announcement, but Anthropic does act on them

The floor on claude-sonnet-4-5-20250929 read “Not sooner than September 29, 2026” for months. On September 30 it became a deprecation with a firm retirement of November 30, 2026, which is 60 days from today. So stop reading a floor as a comfort. Two Active snapshots carry one inside the same window: claude-haiku-4-5-20251001 not sooner than October 15, 2026, which is 14 days out, and claude-opus-4-5-20251101 not sooner than November 24, 2026, 54 days out. Neither has a deprecation notice attached yet. Haiku 4.5 is still one of four models in the current models-overview comparison table, which argues against imminent retirement, but that’s inference and not evidence. Re-check the deprecations page and the API release notes after October 15.

Python SDK v1.0 will stop a build

The August 20, 2026 release moves httpx to httpx2, requires Python 3.10 or later, removes Text Completions and removes temperature, top_p and top_k from Messages. An unpinned dependency will pick this up on its own schedule, which is rarely yours.

The Workbench URL is dead

The legacy Workbench at platform.claude.com/workbench sunset on August 17, 2026 and the replacement Playground at platform.claude.com/playground was documented in the August 18, 2026 release note. Three experimental /v1/experimental/*_prompt endpoints retired on the same August 17 date. Internal links, onboarding docs and screenshots will still point at the old surface.

July 24 breaking changes still apply

Setting thinking to disabled is only valid at effort high or below. Combine it with extra high or max and you get a 400. Requesting fast speed on Opus 4.7 errors. Both break a working integration quietly, right up until a call fails.

A default moved underneath Pro and Team Standard budgets

Claude Code changed the default model on Pro and Team Standard plans from Sonnet to Opus, matching Max, Team Premium and Enterprise. It also made Opus 5.5 the default Opus. If you sized Pro or Team Standard Claude Code usage against a Sonnet default, that assumption expired without anyone telling the budget holder. More broadly, the July 24 announcement put Opus 5 as the Claude Max default and left Sonnet 5 on Free and Pro. No current page restates either at version granularity. Pin a model id in anything you evaluate, or your baseline changes underneath the test. And read the hostname that answers. anthropic.com/pricing returns a 406 and can’t be fetched at all. claude.com/pricing renders per-plan availability in detail and now files Sonnet 5, Opus 5 and Opus 4.8 under “Legacy models” in plain text.

The safety fallback has no published name, and two categories rather than one

Anthropic says dual-use cybersecurity work is redirected “to our Opus models,” plural and unnamed, naming penetration testing, exploit generation and binary-based vulnerability scanning as examples. It also says queries related to research and development in the life sciences are directed to the Opus models. No first-party page names a specific fallback model. Two things follow. If a runbook, incident playbook or user-facing message in your product names one, it’s repeating a guess rather than a documented fact. And if your safety documentation describes routing as a cybersecurity mechanism only, it will surprise a research team the first time a life sciences query behaves differently from the rest.

Stolen API keys and clients spoofing Claude Code

Treat an API key as a production credential, not a convenience. A stolen key lets an attacker run operations at your expense while the activity is attributed to you. That’s a cost exposure and an attribution problem in the same incident review. Anthropic’s September 10, 2026 threat intelligence report also describes credential harvesting through fake Claude resellers, and malicious clients that spoof “Claude Code.” That second one is a supply-chain warning about the tool developers install most casually. So name it in an internal advisory: install Claude Code from Anthropic’s own distribution and nowhere else.

Records built on the Compliance API Activity Feed changed shape retroactively

This is the rare vendor change that reaches backward into data you already collected. From September 24, 2026 the Compliance API Activity Feed no longer returns file names, project document names or artifact titles. The filename and title fields are always empty or omitted, including on activities recorded before the change. So a retention or e-discovery process that passed its test in September can produce different output today against the same period, with no error to tell you. Names must now be resolved by ID with a Compliance Access Key carrying read:compliance_user_data. Re-test any records workflow that reads those fields, and re-test it against historical activities, not just new ones.

The anthropic-version header on Enterprise endpoints

Check what your client actually sends, not what your notes say it sends. The September 1 release notes say the Admin API Claude Enterprise endpoints, the Analytics API and the Compliance API guides now show the anthropic-version header. They instruct you to send it on every request to these endpoints. An integration built against the August 26 Compliance API release may not be sending it. This is a rolling release-note item.

Claude Code claims can no longer be dated from a first-party page

Anthropic’s own documentation stopped carrying this. platform.claude.com/docs/en/release-notes/claude-code now redirects off the documentation domain to the changelog on GitHub, and that file is numbered by version with no dates at all. You can order versions against the Opus 5.5 and Sonnet 5.5 launches, and that’s as close to a calendar as you get. So every Claude Code fact above, the managed-settings controls included, is version evidence rather than dated evidence. Check each against the version you have installed before it enters a control description. Three other items rest on rolling pages rather than dated articles: the self-serve HIPAA item, the August 17 Workbench sunset with the August 18 Playground rename, and the Python SDK v1.0 release.

Vendor benchmark claims, and there are more of them now

Anthropic made a lot of claims in nine days and none of them has been independently checked. Opus 5.5 performs at the level of Fable 5.1 on most work, costs 40 percent less to run than Opus 5 and produces output more than 30 percent faster. Sonnet 5.5 is more than 30 percent faster than Sonnet 5 and costs up to 30 percent less for most work. Fable 5.1 and Mythos 5.1 are the world’s most advanced models for coding and knowledge work. Opus 5 sits within 0.5% of Fable 5 on the benchmark Anthropic cites. Every one of those is the vendor marking its own homework. The Opus 5.5 claim is the expensive one to get wrong, because a whole tiering decision now hangs off it. Run a task-level evaluation before you move production work up or down a price tier on the strength of an adjective.

Environment	Useful boundary	What still requires review
Claude chat	Ephemeral server-side container	Files produced, connector scope and claims
Claude Code	Local sandbox plus human command approvals	Shell commands, diffs, secrets and outbound access
Claude Cowork	Sealed VM and selected folder mounts	Mounted workspace, MCP servers and completed actions
Claude in Chrome	The browser tab and the signed-in session it already holds	Page content as untrusted input, form submissions, and what the session can reach
Claude Tag	Channel-scoped tools, memory and spend	Public responses, tool actions and cross-team access

9 / Sources

Where these facts come from

Product status, availability and pricing move fast. Below are the dated first-party articles behind the facts above, read October 1, 2026. Several load-bearing facts sit on living pages instead: the retention documentation, the Covered Models article, the pricing page, the deprecations table, the models overview and the rolling API release notes. Those pages carry no publication date, so they’re attributed inline rather than listed. The cards after the list say which facts rest on which page, and where Anthropic’s own pages disagree.

Retention rests on living pages

The per-workspace 30-day toggle, the two-year flagged-content figure and the three six-year tails come from Anthropic’s API and data retention documentation. The Covered Models list comes from a support article. Neither page carries a publication date. Read both before a control description quotes them.

Prices, floors and the lineup are readings

The plan and seat structure, the “Legacy models” heading and every legacy price come from the Claude pricing page. Every deprecation status and retirement floor comes from the model deprecations table. The four-model comparison, the Opus 5.5 recommendation and the migration pointer come from the models overview. All three are rewritten in place.

Operational changes sit on rolling release notes

The September 22 and 28 launch entries and their breaking changes. The September 30 Sonnet 4.5 deprecation. The September 24 refusal billing and both Compliance API changes. Cache diagnostics on September 23. The anthropic-version header, the compaction beta, the Managed Agents auto policies, the Workbench sunset, Python SDK v1.0, the three betas and the Sonnet 5 price confirmation. Anthropic rewrites those notes in place, so none of them can anchor a date.

Claude Code items carry no dates

Anthropic’s Claude Code release-notes address, platform.claude.com/docs/en/release-notes/claude-code, now redirects to a changelog on GitHub. That file is numbered by version and carries no dates. So the managed-settings controls, the Bedrock and Vertex downgrade and the Pro and Team Standard default change can be ordered against the two September launches, not dated. The ant CLI items sit in the same place. Check each against your installed version.

Where Anthropic’s pages disagree

Three Anthropic sources still frame Mythos 5.1 access three ways: the Cyber Verification Program, the Life Sciences Verification Program and Project Glasswing. The news pages and the API release notes disagree about interim zero data retention on Fable 5.1, and no page publishes the eligibility criteria. On Opus 5.5, three first-party pages say zero data retention is available, and one reading of the rolling release-notes page appeared to say otherwise. Confirm that one on the release-notes page.

One figure rests on a Microsoft page

The seven-year classification-score retention and Microsoft’s subprocessor terms for Fable 5.1 come from “Anthropic models in Microsoft Online Services” on Microsoft Learn, stamped September 18, 2026 and read October 1, 2026. No Anthropic retention page carries the seven-year figure. Confirm it before it goes near a control document.

When a dated article and a live page disagree

A dated article is authoritative for what was announced and when. A live page is authoritative for what is true today. When they disagree, the live page governs the present and the article governs the record. Keep your own dated copy of the live page, because it carries no history.
· Example: the June 30, 2026 Sonnet 5 article records $2 / $10 as introductory pricing. The live pricing page governs today, and it shows $2 / $10 as the standard price.

Evidence note

The workplace positioning and the behavior guidance are AI Mindset interpretation. Model names, dates, availability, prices and product capabilities come from the linked first-party material. Where a fact lives only on a living reference page, it’s attributed inline with an instruction to confirm it directly. Where Anthropic’s own pages disagree, the disagreement is reported above rather than resolved for you. They currently disagree on three things. How Mythos 5.1 access is gated. Whether eligible customers can use Fable 5.1 under zero data retention. And whether EFS has started rolling out or is about to. One more gets a verdict rather than staying open. On zero data retention for Opus 5.5, three first-party pages say yes and one reading of a rolling page appeared to say no. So treat it as available and check the release-notes page. One disagreement isn’t Anthropic’s alone, and it gets the same treatment. Anthropic’s direct terms and the subprocessor terms Microsoft publishes for Fable 5.1 don’t describe the same retention outcome, and the route you buy through changes the answer. That’s recorded rather than settled in favor of the version that reads better.

AI Mindset