api-20260923T183720Z / page 11 of 11 / documentation capture

about-claude/model-deprecations Changed · +0 / -4 lines

from line 64

6464

6565## Model status

6666

67<Note>

68 [Claude Mythos Preview](https://anthropic.com/glasswing) (`claude-mythos-preview`) is deprecated. To migrate to [Claude Mythos 5](https://anthropic.com/glasswing) (`claude-mythos-5`), see the [migration guide](https://platform.claude.com/docs/en/models/fable-5/migration-guide#migrating-from-claude-mythos-preview).

69</Note>

70

7167Current and recently retired models are listed in the following table with their status:

7268

about-claude/models/choosing-a-model Changed · +0 / -4 lines

from line 107

107107 The latest Opus model: breaking changes, new features, and behavior differences

108108 </Card>

109109

110 <Card title="What's new in Claude Opus 5" icon="sparkle" href="https://platform.claude.com/docs/en/models/opus-5/whats-new-opus-5">

111 Explore the improvements in Claude Opus 5

112 </Card>

113

114110 <Card title="What's new in Claude Sonnet 5" icon="sparkle" href="https://platform.claude.com/docs/en/models/sonnet-5/whats-new-sonnet-5">

115111 For everyday workloads that balance speed and capability

116112 </Card>

agents-and-tools/tool-use/tool-runner Changed · +1 / -1 lines

from line 1093

10931093

10941094### Automatic context management

10951095

1096For long-running agentic tasks, the TypeScript and Ruby tool runners support automatic [compaction](https://platform.claude.com/docs/en/build-with-claude/context-editing#client-side-compaction-sdk), which generates summaries when token usage exceeds a threshold so the conversation can continue beyond context window limits. Both SDKs have deprecated this client-side option in favor of [server-side compaction](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold), which works with every SDK's tool runner through the `context_management` request parameter. The Python SDK (v1.0 and later) and the Go, Java, C#, and PHP tool runners don't include client-side compaction. The Python, TypeScript, C#, Go, and Java tool runners have a `compact_before_next_turn()` helper for on-demand compaction, spelled in each language's own casing. See [Compact in a loop](https://platform.claude.com/docs/en/build-with-claude/compaction-on-demand#compact-in-a-loop). Use it or a `context_management` compaction edit on a runner, not both.

1096For long-running agentic tasks, the TypeScript and Ruby tool runners support automatic [compaction](https://platform.claude.com/docs/en/build-with-claude/context-editing#client-side-compaction-sdk), which generates summaries when token usage exceeds a threshold so the conversation can continue beyond context window limits. Both SDKs have deprecated this client-side option in favor of [server-side compaction](https://platform.claude.com/docs/en/build-with-claude/compaction-threshold), which works with every SDK's tool runner through the `context_management` request parameter. The Python SDK (v1.0 and later) and the Go, Java, C#, and PHP tool runners don't include client-side compaction. The Python, TypeScript, C#, Go, Java, PHP, and Ruby tool runners have a `compact_before_next_turn()` helper for on-demand compaction, spelled in each language's own casing. See [Compact in a loop](https://platform.claude.com/docs/en/build-with-claude/compaction-on-demand#compact-in-a-loop). Use it or a `context_management` compaction edit on a runner, not both.

10971097

10981098### Debugging tool execution

10991099

build-with-claude/batch-processing Changed · +4 / -0 lines

from line 102

102102| Claude Haiku 4.5 | $0.50 / MTok | $2.50 / MTok |

103103| Claude Haiku 3.5 ([retired, except on Bedrock and Google Cloud](https://platform.claude.com/docs/en/about-claude/model-deprecations)) | $0.40 / MTok | $2 / MTok |

104104

105* **MTok:** Million tokens. $5 / MTok is $5 for every million tokens.

106* **Limited access:** Offered separately, by invitation only, as part of [Project Glasswing](https://anthropic.com/glasswing). For access, contact your Anthropic, AWS, or Google Cloud account team.

107* **Retired:** May still be available on other cloud platforms. See [Model deprecations](https://platform.claude.com/docs/en/about-claude/model-deprecations) for more.

108

105109## How to use the Message Batches API

106110

107111### Prepare and create your batch

build-with-claude/claude-on-vertex-ai Changed · +2 / -0 lines

from line 118

118118| Model | Agent Platform API model ID |

119119| :--------------------------------------------------------------------------------------------------- | :--------------------------- |

120120| Claude Fable 5.1 | `claude-fable-5-1` |

121| Claude Mythos 5.1 ([limited availability](https://anthropic.com/glasswing)) | `claude-mythos-5-1` |

121122| Claude Fable 5 | `claude-fable-5` |

123| Claude Mythos 5 ([limited availability](https://anthropic.com/glasswing)) | `claude-mythos-5` |

122124| Claude Opus 5.5 | `claude-opus-5-5` |

123125| Claude Opus 5 | `claude-opus-5` |

124126| Claude Opus 4.8 | `claude-opus-4-8` |

build-with-claude/extended-thinking Changed · +1 / -1 lines

from line 354

354354You need to migrate off `type: "enabled"` if:

355355

356356* You use Claude Opus 4.6 or Claude Sonnet 4.6, where `budget_tokens` is deprecated.

357* You are moving to Claude Opus 4.7, Claude Opus 4.8, Claude Opus 5, Claude Opus 5.5, Claude Sonnet 5, Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, or Claude Mythos 5, where `type: "enabled"` returns a 400 error.

357* You use Claude 4.7 or a later model, such as Claude Opus 5.5, Claude Sonnet 5, or Claude Fable 5.1, where `type: "enabled"` returns a 400 error.

358358

359359The mapping is small: remove `budget_tokens`, set `thinking: {type: "adaptive"}`, and control reasoning depth with `output_config: {effort: ...}` instead of a token budget.

360360

build-with-claude/fast-mode Changed · +2 / -2 lines

from line 27

2727</Note>

2828

2929<Note>

30 Fast mode is not available on Claude Opus 4.7. Requests to `claude-opus-4-7` with `speed: "fast"` return an error; unlike Claude Opus 4.6 (see the following note), requests do not fall back to standard speed. The model itself remains available at standard speed. To continue using fast mode, migrate to Claude Opus 5.5, [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/migration-guide#migrating-from-claude-opus-47), or Claude Opus 4.8.

30 Fast mode is not available on Claude Opus 4.7. Requests to `claude-opus-4-7` with `speed: "fast"` return an error; unlike Claude Opus 4.6 (see the following note), requests do not fall back to standard speed. The model itself remains available at standard speed. To continue using fast mode, see [Migrating to Claude Opus 5.5 from Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#migrating-from-claude-opus-47).

3131</Note>

3232

3333<Note>

34 Fast mode is not available on Claude Opus 4.6. Requests to `claude-opus-4-6` with `speed: "fast"` do not return an error: they run at standard speed and are billed at [standard rates](https://platform.claude.com/docs/en/about-claude/pricing) rather than fast mode's premium rates, and the response reports [`usage.speed: "standard"`](https://platform.claude.com/docs/en/build-with-claude/fast-mode#checking-which-speed-was-used). To continue using fast mode, migrate to Claude Opus 5.5, [Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/migration-guide#migrating-from-claude-opus-46), or Claude Opus 4.8.

34 Fast mode is not available on Claude Opus 4.6. Requests to `claude-opus-4-6` with `speed: "fast"` do not return an error: they run at standard speed and are billed at [standard rates](https://platform.claude.com/docs/en/about-claude/pricing) rather than fast mode's premium rates, and the response reports [`usage.speed: "standard"`](https://platform.claude.com/docs/en/build-with-claude/fast-mode#checking-which-speed-was-used). To continue using fast mode, see [Migrating to Claude Opus 5.5 from Claude Opus 4.6 and earlier Opus models](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#migrating-from-claude-opus-46).

3535</Note>

3636

3737## How fast mode works

build-with-claude/prompt-caching Changed · +2 / -2 lines

from line 666

666666| **Thinking parameters** | Model-specific | Model-specific | ✘ | The thinking configuration (mode, and `budget_tokens` in extended mode) is rendered into the prompt, so changing it always invalidates message blocks; tool and system caches are also invalidated on models that render the configuration ahead of them. See [Thinking and prompt caching](https://platform.claude.com/docs/en/build-with-claude/thinking#thinking-and-prompt-caching). |

667667| **Effort setting** | Model-specific | Model-specific | ✘ | Changing the [`output_config.effort`](https://platform.claude.com/docs/en/build-with-claude/effort) value always invalidates message blocks, with the same model-specific effect on tool and system caches as thinking parameters. Setting effort explicitly to the model's default is equivalent to omitting it and does not invalidate. On models that support [per-message effort](https://platform.claude.com/docs/en/build-with-claude/effort#change-effort-mid-conversation-beta), an effort change carried in a `role: "system"` message inside `messages` leaves the cached prefix intact. |

668668| **Non-tool results passed to extended thinking requests** | ✓ | ✓ | Model-specific | On Opus 4.5+ and Sonnet 4.6+, thinking blocks are preserved by default, so the cache remains valid (✓). On earlier Opus/Sonnet models and all Haiku models, all previously-cached thinking blocks are stripped from context, and any messages that follow those thinking blocks are removed from the cache (✘). For more details, see [Caching with thinking blocks](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#caching-with-thinking-blocks). |

669| **Dropped thinking blocks** | ✓ | ✓ | ✘ | When the API drops a Claude Fable 5.1 or Claude Mythos 5.1 thinking block that isn't [preserved](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-thinking) on that request (for example, one you replay to an earlier model), the cached prefix changes from that block's position onward on that request. Blocks the receiving model can read, passed back unchanged, keep the cache intact. |

669| **Dropped thinking blocks** | ✓ | ✓ | ✘ | When the API drops a Claude Fable 5.1, Claude Mythos 5.1, or Claude Opus 5.5 thinking block that isn't [preserved](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-thinking) on that request (for example, one you replay to a model that can't read it), the cached prefix changes from that block's position onward on that request. Blocks the receiving model can read, passed back unchanged, keep the cache intact. |

670670

671671On models that support [mid-conversation tool changes](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages#mid-conversation-tool-changes), the `inline-tools-2026-09-15` beta header lets you add a tool, or change a tool's definition, partway through a conversation without editing `tools`. Send the definition in a `tool_addition` block in a mid-conversation system message and leave `tools` exactly as you first sent it. The cached prefix still matches, so only the appended message is processed as new input. The one exception is a `tools` array with no non-deferred tool, where the first tool defined this way costs one full cache miss on that request. See [Define tools in a message](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages#define-tools-in-a-message-beta).

672672

from line 1066

10661066 messages: [['role' => 'user', 'content' => 'warmup']],

10671067 );

10681068

1069 echo $prewarm->stopReason->value, PHP_EOL; // "max_tokens"

1069 echo $prewarm->stopReason, PHP_EOL; // "max_tokens"

10701070 echo json_encode($prewarm->content), PHP_EOL; // []

10711071 echo json_encode($prewarm->usage), PHP_EOL;

10721072 ```

build-with-claude/prompt-engineering/claude-prompting-best-practices Changed · +2 / -2 lines

from line 11

1111* **Migration considerations** last, for prompts moving from earlier generations.

1212

1313<Tip>

14 For an overview of model capabilities, see the [models overview](https://platform.claude.com/docs/en/models/overview). For Claude Fable 5.1 capabilities and API changes, see [What's new in Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1). For Claude Fable 5 capabilities and API changes, see [Introducing Claude Fable 5 and Claude Mythos 5](https://platform.claude.com/docs/en/models/fable-5/introducing-claude-fable-5-and-claude-mythos-5). For details on what's new in Claude Sonnet 5, see [What's new in Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/whats-new-sonnet-5). For details on what's new in Claude Opus 5, see [What's new in Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/whats-new-opus-5). For migration guidance, see the [Migration guide](https://platform.claude.com/docs/en/about-claude/models/migration-guide). For Claude Opus 5.5, see [What's new in Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/whats-new-opus-5-5).

14 For an overview of model capabilities, see the [models overview](https://platform.claude.com/docs/en/models/overview). For Claude Fable 5.1 capabilities and API changes, see [What's new in Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/whats-new-fable-5-1). For Claude Fable 5 capabilities and API changes, see [Introducing Claude Fable 5 and Claude Mythos 5](https://platform.claude.com/docs/en/models/fable-5/introducing-claude-fable-5-and-claude-mythos-5). For details on what's new in Claude Sonnet 5, see [What's new in Claude Sonnet 5](https://platform.claude.com/docs/en/models/sonnet-5/whats-new-sonnet-5). For migration guidance, see the [Migration guide](https://platform.claude.com/docs/en/about-claude/models/migration-guide). For Claude Opus 5.5, see [What's new in Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/whats-new-opus-5-5).

1515</Tip>

1616

1717## Model-specific guidance

from line 777

777777* **Prefer general instructions over prescriptive steps.** A prompt like "think thoroughly" often produces better reasoning than a hand-written step-by-step plan. Claude's reasoning frequently exceeds what a human would prescribe.

778778* **Multishot examples work with thinking.** Use `<thinking>` tags inside your few-shot examples to show Claude the reasoning pattern. It will generalize that style to its own extended thinking blocks.

779779* **Manual chain-of-thought (CoT) prompting as a fallback.** When thinking is off, you can still encourage step-by-step reasoning by asking Claude to think through the problem. Use structured tags like `<thinking>` and `<answer>` to cleanly separate reasoning from the final output. On Claude Opus 5, prefer keeping thinking enabled at a lower effort level instead: with thinking disabled, the model can occasionally emit internal XML tags into its visible output, so see [Running with thinking disabled](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5#running-with-thinking-disabled) before applying this pattern there.

780* **Ask Claude to self-check.** Append something like "Before you finish, verify your answer against \[test criteria]." This catches errors reliably, especially for coding and math. Claude Opus 5 is the exception: it verifies its own work well without explicit instruction, and verification instructions carried over from prompts tuned for earlier models can cause over-verification, adding tokens and latency. When migrating to Claude Opus 5, remove these instructions rather than rewriting them. See [Task scope and over-verification](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5#task-scope-and-over-verification).

780* **Ask Claude to self-check.** Append something like "Before you finish, verify your answer against \[test criteria]." This catches errors reliably, especially for coding and math. Claude Opus 5 is the exception: it verifies its own work well without explicit instruction, and verification instructions carried over from prompts tuned for earlier models can cause over-verification, adding tokens and latency. On Claude Opus 5, remove these instructions rather than rewriting them. See [Task scope and over-verification](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5#task-scope-and-over-verification).

781781

782782<Note>

783783 When extended thinking is disabled, Claude Opus 4.5 is particularly sensitive to the word "think" and its variants. Consider using alternatives like "consider," "evaluate," or "reason through" in those cases.

build-with-claude/prompt-engineering/prompting-claude-fable-5-1 Changed · +1 / -1 lines

from line 761

761761

762762## Writing density

763763

764Claude Fable 5.1's writing is generally a step up from earlier Claude models, with fewer stock phrases and less unexplained jargon. In some cases, though, its prose is denser than Claude Fable 5's: sentences run longer and there are fewer paragraph breaks. An instruction that defines the anti-pattern, mannered prose, helps. Add it to a user message (preferred) or the system prompt:

764Claude Fable 5.1's writing has few stock phrases and little unexplained jargon. In some cases, though, its prose is denser than Claude Fable 5's: sentences run longer and there are fewer paragraph breaks. An instruction that defines the anti-pattern, mannered prose, helps. Add it to a user message (preferred) or the system prompt:

765765

766766```text wrap

767767Mannered prose substitutes metaphor and flourish for direct statement. Instead of "a parameter worth varying," the mannered writer produces "a dial worth turning." Instead of "this point still matters," they write "this point earns its keep." The phrases exist to display the writer, not to convey the idea, and readers can tell. That is why mannered prose irritates: it makes the reader work harder so the writer can perform. It is also imprecise. Metaphors drag in connotations the writer did not choose and cannot control. The fix is to say what you mean. When a literal phrase is available, use it.

build-with-claude/prompt-engineering/prompting-claude-opus-4-8 Changed · +2 / -2 lines

from line 4

44description: Behavioral differences and prompting patterns for Claude Opus 4.8, covering verbosity, effort calibration, tool use, subagents, and frontend defaults.

55---

66

7This guide covers the prompting patterns specific to Claude Opus 4.8. For the API changes involved in moving from Claude Opus 4.8 to the latest Opus model, see [Migrating to Claude Opus 5 from Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-5/migration-guide#migrating-from-claude-opus-4-8-to-claude-opus-5). For techniques that apply across all current Claude models, see [Prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).

7This guide covers the prompting patterns specific to Claude Opus 4.8. For the API changes involved in moving from Claude Opus 4.8 to Claude Opus 5.5, see [Migrating to Claude Opus 5.5 from Claude Opus 4.8](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#migrating-from-claude-opus-4-8). For techniques that apply across all current Claude models, see [Prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices).

88

99Claude Opus 4.8 has particular strengths in long-horizon agentic work, knowledge work, vision, and memory tasks. It performs well out of the box on existing Claude Opus 4.7 prompts. The following patterns cover the behaviors that most often require tuning.

1010

1111<Note>

12 For the API parameter changes since Claude Opus 4.7 (sampling parameters, effort default, 1M context window default, mid-conversation system messages, and refusal stop details), see [Migrating to Claude Opus 5 from Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-5/migration-guide#migrating-from-claude-opus-47), which covers the same changes on the way to the latest Opus model; Claude Opus 4.8 shares these behaviors.

12 For the API parameter changes since Claude Opus 4.7 (1M context window default and mid-conversation system messages), see [Migrating to Claude Opus 5.5 from Claude Opus 4.7](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#migrating-from-claude-opus-47), which covers the same changes on the way to Claude Opus 5.5; Claude Opus 4.8 shares these behaviors.

1313</Note>

1414

1515## Response length and verbosity

build-with-claude/refusals-and-fallback Changed · +2 / -2 lines

from line 1107

11071107

11081108* Retries walk your fallback list in order. A fallback model that itself refuses passes the request to the next entry.

11091109* When every model in the list has declined, the middleware returns the final refusal (the last model's refusal response) rather than raising an error.

1110* Thinking blocks from Claude Fable 5.1 or Claude Fable 5 pass through unchanged. Each retry re-sends your original request body, and the only blocks the middleware removes from conversation history on later requests are the `fallback` boundary blocks it added itself. The fallback model can't read Claude Fable 5.1 blocks, which are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them.

1110* Thinking blocks from Claude Fable 5.1, Claude Opus 5.5, or Claude Fable 5 pass through unchanged. Each retry re-sends your original request body, and the only blocks the middleware removes from conversation history on later requests are the `fallback` boundary blocks it added itself. The fallback model can't read Claude Fable 5.1 blocks, which are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them. The API also drops Claude Opus 5.5 blocks for every fallback model except Claude Fable 5.1 and Claude Mythos 5.1 (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)).

11111111* Responses served through the middleware include a `fallback` content block at each model boundary, the same as server-side fallback responses. The middleware manages those blocks for you on later requests.

11121112* The model that accepted is recorded in `BetaFallbackState`, so follow-up requests that share the state stay pinned to it rather than re-asking a model that refused.

11131113

from line 1127

11271127 <Step title="Re-send on a fallback model">

11281128 Send the same request with `model` set to a fallback model, such as Claude Opus 4.8. Another model can normally serve a request that Claude Fable 5.1 or Claude Fable 5 declines. How you handle the conversation history depends on whether you redeem a [fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit):

11291129

1130 * **Not redeeming a credit:** you can leave the earlier `thinking` and `redacted_thinking` blocks in place or strip them to save input tokens. The fallback model cannot use them either way: it ignores Claude Fable 5 blocks, and Claude Fable 5.1 blocks are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them.

1130 * **Not redeeming a credit:** you can leave the earlier `thinking` and `redacted_thinking` blocks in place or strip them to save input tokens. The fallback model normally can't use them either way: it ignores Claude Fable 5 blocks, and Claude Fable 5.1 blocks are [preserved only for that model or a newer one](https://platform.claude.com/docs/en/build-with-claude/thinking#preserved-for-model), so the API drops them. The API also drops Claude Opus 5.5 blocks for every fallback model except Claude Fable 5.1 and Claude Mythos 5.1 (see [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)).

11311131 * **Redeeming a credit:** send the body unchanged, because redemption requires an exact match. The server handles the earlier model's thinking blocks on a redemption, so do not strip them (see [Fields that must match the refused request](https://platform.claude.com/docs/en/build-with-claude/fallback-credit#reference)).

11321132 </Step>

11331133

build-with-claude/working-with-messages Changed · +1 / -1 lines

from line 20

2020## Basic request and response

2121

2222<Note>

23 The `temperature`, `top_p`, and `top_k` sampling parameters are not supported on Claude 4.7 and later models and Claude Mythos Preview. Setting them to a non-default value returns a 400 error. Omit them from request payloads and use prompting to guide the model's behavior instead. See the [migration guide](https://platform.claude.com/docs/en/models/opus-5/migration-guide#migrating-from-claude-opus-47).

23 The `temperature`, `top_p`, and `top_k` sampling parameters are not supported on Claude 4.7 and later models and Claude Mythos Preview. Setting them to a non-default value returns a 400 error. Omit them from request payloads and use prompting to guide the model's behavior instead. See the [migration guide](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#opus-46-breaking-changes).

2424</Note>

2525

2626<CodeGroup>

claude_api_primer Changed · +1 / -0 lines

from line 243

243243

244244Thinking is supported in the following models:

245245

246* Claude Opus 5.5 (`claude-opus-5-5`, adaptive thinking only, always on)

246247* Claude Opus 5 (claude-opus-5, adaptive thinking only, on by default)

247248* Claude Sonnet 5 (`claude-sonnet-5`, adaptive thinking only, on by default)

248249* Claude Opus 4.8 (claude-opus-4-8, adaptive thinking only)

manage-claude/compliance-content-data Changed · +2 / -2 lines

from line 24

2424

2525Use [List chats](https://platform.claude.com/docs/en/api/compliance/apps/chats/list) to page through chat metadata, then [Get chat messages](https://platform.claude.com/docs/en/api/compliance/apps/chats/messages/list) to fetch the full message content of one chat.

2626

27The chat list endpoint defaults to organization-wide scope: leave off `user_ids[]` to include every chat under your parent organization. Add `order_by=updated_at` to sort by last update time. This combination is the recommended way to export chats and keep an export current, because one paginated loop picks up new chats, modified chats, and chats deleted in claude.ai for every user without enumerating users first. The following request lists chats updated since a given date.

27The chat list endpoint defaults to organization-wide scope: leave off `user_ids[]` to include every chat under your parent organization. Add `order_by=updated_at` to sort by last update time. This combination is the recommended way to export chats and keep an export current, because one paginated loop picks up new chats, chats with new messages, and chats deleted in claude.ai for every user without enumerating users first. The following request lists chats updated since a given date.

2828

2929```bash cURL

3030curl --fail-with-body -sS -G \

from line 63

6363

6464Results sort ascending by the `order_by` field, oldest first, with ties broken by `id`. Pagination uses the standard `first_id`/`last_id`/`has_more` cursor fields described in [Paginate results](https://platform.claude.com/docs/en/manage-claude/compliance-activity-feed#paginate-results). To walk forward toward newer chats, pass the response's `last_id` back as `after_id` on the next request.

6565

66That forward walk is also how you keep an export current across runs: persist the final page's `last_id` and resume from it as `after_id` on the next run. Because the list is ordered by `updated_at`, a chat that changes after your saved cursor reappears ahead of it, so each incremental run returns both brand-new chats and older chats that have since been modified or deleted in claude.ai. Process results idempotently, keyed by chat `id`, to handle those reappearances. A chat that comes back with `deleted_at` populated has no content left to fetch, so treat it as deleted rather than updated.

66That forward walk is also how you keep an export current across runs: persist the final page's `last_id` and resume from it as `after_id` on the next run. Because the list is ordered by `updated_at`, a chat reappears ahead of your saved cursor when it receives a new message, is moved into or out of a project, or is deleted in claude.ai. Each incremental run therefore returns both brand-new chats and older chats that have since changed in one of those ways. Other edits, such as a rename, are not guaranteed to make a chat reappear. Process results idempotently, keyed by chat `id`, to handle those reappearances. A chat that comes back with `deleted_at` populated has no content left to fetch, so treat it as deleted rather than updated.

6767

6868A few constraints apply to these organization-wide queries. Cursors are opaque and bound to the sort key, so an `after_id` issued under one `order_by` value is rejected with a 400 error under the other. Time-filter bounds must match the sort key too: pair `updated_at.*` bounds with `order_by=updated_at`, and `created_at.*` bounds with the default `order_by=created_at`. Backward pagination with `before_id` is not supported, and the `project_ids[]` filter is not available. See [List chats](https://platform.claude.com/docs/en/api/compliance/apps/chats/list) for the full filter reference.

6969

manage-claude/compliance-sessions Changed · +2 / -0 lines

from line 278

278278

279279Two parameters cap how many bytes of each tool block are returned: `tool_use_input_max_bytes` and `tool_result_max_bytes`, both defaulting to 10,000 bytes. Pass `-1` for the server maximum (about 1 MiB per string); `0` returns [400 Bad Request](https://platform.claude.com/docs/en/manage-claude/compliance-errors#400-bad-request), and values above the maximum are clamped to it. A string cut off by either cap is cut on a character boundary and has an in-band suffix appended (for example, `…[truncated; pass tool_result_max_bytes=-1 for the server max]`), and its block carries `"truncated": true`. A truncated `tool_use` `input` is therefore no longer valid JSON, so parse tool inputs only from untruncated blocks (or raise the cap and refetch). Blocks of type `text` are always capped at the same server maximum of about 1 MiB; no parameter raises it, and a `text` block at the bound also carries `"truncated": true`.

280280

281Claude Science calls connectors (MCP servers) from code that it runs through its `repl` tool, not as separately named tools, so no block in a Claude Science transcript is named after a connector. Each connector call appears in the code inside a `repl` `tool_use` block's `input` (for example, a `host.mcp("<server>", "<tool>", ...)` call), and connector output appears in the matching `tool_result` only where that code printed it. Cowork and Claude Code sessions differ: they call each connector tool under its own `mcp__<server>__<tool>` name, which is the `name` of the `tool_use` block. To monitor connector use in Claude Science sessions, parse the `input` string and match on the code it contains rather than on a tool name. Pass `tool_use_input_max_bytes=-1` for these sessions so that a long code input is returned up to the server maximum instead of being cut at the 10,000-byte default before the connector call appears.

282

281283Transcript content honors the retention period described under [Sessions on users' machines](https://platform.claude.com/docs/en/manage-claude/compliance-sessions#retrieve-local-sessions). When the start of a session has aged past it, the transcript begins with a single `content_unavailable` placeholder with `reason` of `retention_elapsed`, and the retained messages follow. When every call in a session has aged out, the messages endpoint returns [404 Not Found](https://platform.claude.com/docs/en/manage-claude/compliance-errors#404-not-found), as it does for sessions in organizations your key cannot read, sessions that do not exist, and sessions for which zero data retention is in effect. A malformed session ID returns [400 Bad Request](https://platform.claude.com/docs/en/manage-claude/compliance-errors#400-bad-request).

282284

283285## Sessions in the cloud (remote sessions)

models/fable-5/migration-guide Changed · +1 / -1 lines

from line 363

363363## Migrating to Claude Mythos 5 and Claude Fable 5 from Claude Opus 4.8

364364

365365<Note>

366 If your code is on Claude Opus 4.7 or earlier, first apply the relevant [Migrating to Claude Opus 5](https://platform.claude.com/docs/en/models/opus-5/migration-guide) from-section for the API-level changes from your current model, then the remaining delta in this section.

366 If your code is on Claude Opus 4.7 or earlier, first apply the relevant [Migrating to Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/migration-guide) from-section for the API-level changes from your current model, then the remaining delta in this section.

367367</Note>

368368

369369Migration is mostly drop-in. Claude Fable 5 and Claude Mythos 5 use the same [Messages API](https://platform.claude.com/docs/en/build-with-claude/working-with-messages) and the same [tool use](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview) patterns as Claude Opus 4.8, with the same [1M token context window](https://platform.claude.com/docs/en/build-with-claude/context-windows) by default and the same [128k max output tokens](https://platform.claude.com/docs/en/models/overview). Token counts are roughly unchanged because the models use the same tokenizer. The key changes to check are always-on [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking), thinking output, safety classifier refusals (Claude Fable 5 only), and pricing.

models/overview Changed · +1 / -1 lines

from line 51

5151* **Default effort:** The effort parameter’s default on the Claude API. Set effort explicitly to use a different level.

5252* **Context window:** 1M tokens is roughly 555k words or 2.5M Unicode characters on the current tokenizer (introduced with Claude Opus 4.7); models before it fit about 750k words in 1M tokens. 200k tokens is roughly 150k words.

5353* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5, Claude Opus 4.8, Claude Opus 4.7, Claude Opus 4.6, and Claude Sonnet 4.6 support up to 300k output tokens with the output-300k-2026-03-24 beta header.

54* **Reliable knowledge cutoff:** The date through which the model’s knowledge is most extensive and reliable. Training data cutoff (under Show all details) is the broader range of data used. See Anthropic’s Transparency Hub for details.

54* **Reliable knowledge cutoff:** The date through which the model’s knowledge is most extensive and reliable. Training data cutoff (under Additional details) is the broader range of data used. See Anthropic’s Transparency Hub for details.

5555* **Retirement:** Anthropic’s commitment for Anthropic-operated platforms (Claude API, Claude Platform on AWS, Microsoft Foundry). Amazon Bedrock and Google Cloud set their own dates.

5656* **Claude API alias:** For models before the 4.6 generation, the alias is a convenience pointer that resolves to the dated ID. Dateless IDs are their own pinned snapshot; the alias row repeats them.

5757* **Amazon Bedrock ID:** The ID on Bedrock’s Messages-API endpoint (Claude Opus 4.7 and later, plus Claude Haiku 4.5); a model offered only through Bedrock’s InvokeModel integration shows that ID instead. Bedrock offers global endpoints (dynamic routing) and regional endpoints (guaranteed data routing) for Claude Sonnet 4.5 and later, and sets its own lifecycle dates.

One read of Claude Developer Platformapi-20260923T183720Z

What this read moved