api-20261007T190728Z / page 2 of 4 / documentation capture

build-with-claude/refusals-and-fallback Changed · +4 / -4 lines

from line 1

11---

22title: Refusals and fallback

33url: https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback

4description: How Claude Fable models, Claude Opus models, and Claude Sonnet 5.5 return classifier refusals and how to retry refused requests on a fallback model.

4description: How Claude Fable models, Claude Opus models, Claude Sonnet 5.5, and Claude Haiku 5.5 return classifier refusals and how to retry refused requests on a fallback model.

55---

66

7Claude Fable 5.1, Claude Fable 5, Claude Opus 5.5, Claude Opus 5, and Claude Sonnet 5.5 include safety classifiers that can decline a request. When that happens, you receive a normal response, not an error, with `stop_reason: "refusal"`. Its `stop_details.category` names the policy area (see [What a refusal looks like](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#refusal-response)). You can usually still get an answer by sending the same request to another Claude model. This page shows you how to recognize a refusal and how to set up that retry.

7Claude Fable 5.1, Claude Fable 5, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, and Claude Haiku 5.5 include safety classifiers that can decline a request. When that happens, you receive a normal response, not an error, with `stop_reason: "refusal"`. Its `stop_details.category` names the policy area (see [What a refusal looks like](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#refusal-response)). You can usually still get an answer by sending the same request to another Claude model. This page shows you how to recognize a refusal and how to set up that retry.

88

99Read this page when you build on any of these models and want declined requests to fall through to another model automatically. It also applies when you have seen `"refusal"` in a response and want to know what to do next.

1010

from line 232

232232Server-side fallback retries a refused request inside a single API call. In the default mode, when the primary model declines and the refusal category has a recommended fallback, the API runs the same request on the model Anthropic recommends for that category. You can instead [name up to three fallback models of your own](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#naming-your-own-fallback-models). Either way, you get back one response that names the model that answered, so your user gets an answer in one round trip.

233233

234234<Note>

235 Server-side fallback is in beta on the Claude API. The `fallbacks` parameter is not supported on the [Message Batches API](https://platform.claude.com/docs/en/build-with-claude/batch-processing) (a batch item that includes it comes back as an errored result) and is not available on Amazon Bedrock, Google Cloud, or Microsoft Foundry. On those platforms, use [client-side fallback with the SDK middleware](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#client-side-fallback) instead.

235 Server-side fallback is in beta on the Claude API. The `fallbacks` parameter is not supported on the [Message Batches API](https://platform.claude.com/docs/en/build-with-claude/batch-processing) (a batch item that includes it comes back as an errored result) and is not available on Amazon Bedrock, Google Cloud, or Microsoft Foundry. On those platforms, use [client-side fallback with the SDK middleware](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#client-side-fallback) instead. Claude Haiku 5.5 has no server-side fallback: with `fallbacks: "default"`, a declined request stays declined, and a list of fallback models returns a 400 error.

236236</Note>

237237

238238### Making the request

from line 1157

11571157 </Step>

11581158</Steps>

11591159

1160A manual retry writes the fallback model's prompt cache from scratch, which costs more than reading an existing cache. [Fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit) refunds that cost; redeem it on every retry you build yourself.

1160A manual retry writes the fallback model's prompt cache from scratch, which costs more than reading an existing cache. [Fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit) refunds that cost; redeem it on every retry you build yourself. A Claude Haiku 5.5 refusal carries no fallback credit, so a retry after one writes the fallback model's cache at full price.

11611161

11621162## Refusals in Message Batches

11631163

build-with-claude/thinking Changed · +16 / -8 lines

from line 52

5252| --------------------- | ------------------- | ----------------- | -------------------------------- | ----------------------------------------------- | -------------------------------------- |

from line 70

6970

7071In the table, "at `high` effort or below" means the request works at `low`, `medium`, and `high` effort and returns a 400 error at `xhigh` or `max`.

7172

72On Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, and Claude Mythos Preview, thinking is already on and needs no configuration. `display` defaults to `"omitted"` on these models, so the thinking text is hidden until you opt in. Opt in with `thinking: {"type": "adaptive", "display": "summarized"}`, which is exactly the following request with the [model string](https://platform.claude.com/docs/en/models/overview) swapped.

73On Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Mythos Preview, and Claude Haiku 5.5, thinking is already on and needs no configuration. `display` defaults to `"omitted"` on these models, so the thinking text is hidden until you opt in. Opt in with `thinking: {"type": "adaptive", "display": "summarized"}`, which is exactly the following request with the [model string](https://platform.claude.com/docs/en/models/overview) swapped.

7374

7475On Claude Opus 4.8, Claude Opus 4.7, Claude Opus 4.6, and Claude Sonnet 4.6, thinking is off until you set `thinking: {type: "adaptive"}`, which lets Claude determine when and how deeply to think based on the request. The following examples do that, set `display: "summarized"` so the thinking text is visible, and use a roomy `max_tokens`:

7576

from line 469

468469

469470Claude Sonnet 5.5 also has thinking on by default, and it rejects `thinking: {type: "disabled"}` with a 400 error. To turn off up-front thinking, send `thinking: {type: "between_tools"}` instead. It's the lowest thinking setting on Claude Sonnet 5.5, and it's accepted at [effort](https://platform.claude.com/docs/en/build-with-claude/effort) `high` or below. The model still returns its [progress updates between tool calls](https://platform.claude.com/docs/en/build-with-claude/thinking#progress-updates). Without tools, the response contains only text, as with `disabled` on Claude Sonnet 5. See [Running without up-front thinking](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-sonnet-5-5#running-without-up-front-thinking) for prompting guidance.

470471

472Claude Haiku 5.5 also has thinking on by default and accepts `thinking: {type: "disabled"}` at [effort](https://platform.claude.com/docs/en/build-with-claude/effort) `high` or below. At `xhigh` or `max` effort, that combination returns a 400 error. To get less thinking, lower the effort level first. With thinking off, the model can skip a tool call it needs when you also request JSON output. See [Use effort to control thinking](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-haiku-5-5#use-effort-to-control-thinking) and [Use adaptive thinking with JSON output and your own tools](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-haiku-5-5#json-output-with-your-own-tools).

473

471474Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Opus 5.5, and Claude Mythos Preview reject `thinking: {type: "disabled"}`. Thinking can't be turned off on these models.

472475

473476To check whether a model accepts `"disabled"` before you send a request, read its `capabilities.thinking.types.disabled.supported` value from the Models API. [Using the Models API](https://platform.claude.com/docs/en/models/overview#using-the-models-api) describes the field.

from line 484

481484The `display` field on the thinking configuration controls how thinking content is returned in API responses. `display` works in both modes: set it alongside `type: "adaptive"` or `type: "enabled"`. It accepts these values:

482485

483486* `"summarized"`: thinking blocks contain [summarized thinking](https://platform.claude.com/docs/en/build-with-claude/thinking#summarized-thinking) text, a readable summary of Claude's reasoning. This is the default on Claude Opus 4.6, Claude Sonnet 4.6, and earlier models.

484* `"omitted"`: thinking blocks are returned with an empty `thinking` field. The `signature` field still carries the encrypted full thinking for multi-turn continuity (see [Thinking encryption](https://platform.claude.com/docs/en/build-with-claude/thinking#thinking-encryption)). This is the default on Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Opus 4.8, Claude Opus 4.7, and [Claude Mythos Preview](https://anthropic.com/glasswing).

487* `"omitted"`: thinking blocks are returned with an empty `thinking` field. The `signature` field still carries the encrypted full thinking for multi-turn continuity (see [Thinking encryption](https://platform.claude.com/docs/en/build-with-claude/thinking#thinking-encryption)). This is the default on Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Opus 4.8, Claude Opus 4.7, [Claude Mythos Preview](https://anthropic.com/glasswing), and Claude Haiku 5.5.

485488* `"updates"` (beta): reasoning blocks are returned with an empty `thinking` field, as with `"omitted"`, and the short [progress updates](https://platform.claude.com/docs/en/build-with-claude/thinking#progress-updates) some models write between tool calls come back as readable text. Requires the beta header `thinking-display-updates-2026-08-18`.

486489

487490Set `display: "omitted"` when your application doesn't surface thinking content to users. The primary benefit is faster time-to-first-text-token when streaming: the server skips streaming thinking tokens entirely and delivers only the signature, so the final text response begins streaming sooner.

from line 1065

10621065

10631066Whether thinking blocks from previous assistant turns stay in context by default depends on the model:

10641067

1065* **Keep all prior turns:** Claude Opus 4.5 and later Opus models, Claude Sonnet 4.6 and later Sonnet models, Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, and Claude Mythos Preview.

1068* **Keep all prior turns:** Claude Opus 4.5 and later Opus models, Claude Sonnet 4.6 and later Sonnet models, Claude Haiku 5.5, Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, and Claude Mythos Preview.

10661069* **Keep the last turn only:** earlier Opus and Sonnet models, and all Haiku models through Claude Haiku 4.5. When you pass older thinking blocks back, the API strips them automatically. You don't need to remove them yourself.

10671070

10681071Preservation brings two benefits:

from line 1075

10721075

10731076The tradeoff is context usage: long conversations consume more context space on keep-all models, because retained thinking blocks count as input like any other conversation history (see [Thinking and the context window](https://platform.claude.com/docs/en/build-with-claude/thinking#thinking-and-the-context-window)). The behavior is automatic in both regimes. No code changes or beta headers are required, and you should keep passing complete, unmodified thinking blocks back as described in [Preserving thinking blocks](https://platform.claude.com/docs/en/build-with-claude/thinking#preserving-thinking-blocks). To override the default in either direction, use [thinking block clearing](https://platform.claude.com/docs/en/build-with-claude/context-editing#thinking-block-clearing).

10741077

1075**Switching models mid-conversation.** Keep passing thinking blocks back unchanged when you switch models, for example after a [classifier refusal fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback). A thinking block is readable only by the model that produced it and certain other models, and the API ignores or drops the blocks the target model can't read. On Claude Fable 5.1 and Claude Mythos 5.1 the direction matters: they read every earlier model's thinking blocks and no earlier model reads theirs, so switching up to them keeps the conversation's reasoning and switching down drops it (see [how dropped blocks are billed and reported](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). Claude Opus 5.5 reads Claude Opus 5's thinking blocks and those of earlier Opus, Sonnet, and Haiku models, but not those of the Claude Fable and Claude Mythos models; on the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read Claude Opus 5.5's blocks. A switch from Claude Opus 5.5 up to Claude Fable 5.1 on the Claude API keeps the earlier turns' reasoning; a switch from Claude Fable 5.1 to Claude Opus 5.5 drops it. Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet 5, Claude Opus 4.8, Claude Haiku 4.5, and earlier models, but not from Claude Opus 5, Claude Opus 5.5, or any Claude Fable or Claude Mythos model. On the Claude API and Google Cloud, Claude Opus 5.5 reads Claude Sonnet 5.5's blocks and no other model does: a switch from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API and Google Cloud keeps the earlier turns' reasoning, and any other switch away from Claude Sonnet 5.5 drops it. Strip prior `thinking` and `redacted_thinking` blocks yourself only to save input tokens on models that ignore rather than drop them, and never when redeeming a [fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit), which requires the body unchanged.

1078**Switching models mid-conversation.** Keep passing thinking blocks back unchanged when you switch models, for example after a [classifier refusal fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback). A thinking block is readable only by the model that produced it and certain other models, and the API ignores or drops the blocks the target model can't read. On Claude Fable 5.1 and Claude Mythos 5.1 the direction matters: they read every earlier model's thinking blocks and no earlier model reads theirs, so switching up to them keeps the conversation's reasoning and switching down drops it (see [how dropped blocks are billed and reported](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models)). Claude Opus 5.5 reads Claude Opus 5's thinking blocks and those of earlier Opus, Sonnet, and Haiku models, and, on the Claude API and Google Cloud, of Claude Haiku 5.5, but not those of the Claude Fable and Claude Mythos models; on the Claude API, Claude Fable 5.1 and Claude Mythos 5.1 read Claude Opus 5.5's blocks. A switch from Claude Opus 5.5 up to Claude Fable 5.1 on the Claude API keeps the earlier turns' reasoning; a switch from Claude Fable 5.1 to Claude Opus 5.5 drops it. Claude Sonnet 5.5 reads thinking blocks from Claude Sonnet 5, Claude Opus 4.8, Claude Haiku 4.5, and earlier models, and, on the Claude API and Google Cloud, from Claude Haiku 5.5, but not from Claude Opus 5, Claude Opus 5.5, or any Claude Fable or Claude Mythos model. On the Claude API and Google Cloud, Claude Opus 5.5 reads Claude Sonnet 5.5's blocks and no other model does: a switch from Claude Sonnet 5.5 up to Claude Opus 5.5 on the Claude API and Google Cloud keeps the earlier turns' reasoning, and any other switch away from Claude Sonnet 5.5 drops it. Strip prior `thinking` and `redacted_thinking` blocks yourself only to save input tokens on models that ignore rather than drop them, and never when redeeming a [fallback credit](https://platform.claude.com/docs/en/build-with-claude/fallback-credit), which requires the body unchanged.

10761079

10771080## Preserved thinking

10781081

from line 1084

10811084* **The model that produced it.** Each model reads its own thinking blocks and those of a fixed set of other models. Claude Fable 5.1 reads blocks from Claude Opus 5 and, on the Claude API, from Claude Opus 5.5; neither Claude Opus 5 nor Claude Opus 5.5 reads blocks from Claude Fable 5.1. The API drops a block the current model can't read, without an error and without billing it. See [Switching models mid-conversation](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#switching-models).

10821085* **Everything sent before it.** A block stays valid only while the top-level `system` prompt, the `tools`, and the messages before it are unchanged. If any of them changes, that block and every later thinking block are invalid, and the API rejects the request with a 400 error or drops the invalid blocks, whichever you choose. See [Keeping the prefix unchanged](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#prefix-check).

10831086

1084Claude Sonnet 5.5's thinking blocks are also tied to the account that produced them. See [Thinking blocks stay with the account that produced them](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#account-bound-thinking) for where the API enforces this.

1087Thinking blocks from Claude Sonnet 5.5 and Claude Haiku 5.5 are also tied to the account that produced them. See [Thinking blocks stay with the account that produced them](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#account-bound-thinking) for where the API enforces this.

10851088

10861089The model check applies to every account. The API enforces the prefix check by default for accounts created on or after August 31, 2026, 00:00 UTC. On older accounts it enforces the check only on requests that set `thinking.block_binding.prefix_mismatch_behavior`. Make your integration append-only regardless of your account's age, so the same code works on every account, including newer accounts enforced by default.

10871090

from line 1147

11441147

11451148The following diagrams illustrate the last-turn-only (stripping) regime. The first shows a multi-turn conversation: each turn's thinking block is generated in the output but not carried into later turns' input.

11461149

1147![Diagram of thinking on a model that strips previous thinking blocks: each turn's thinking block is generated in the output and not carried into later turns' input](https://platform.claude.com/docs/images/context-window-thinking.svg)

1150<Frame>

1151 ![Diagram of thinking on a model that strips previous thinking blocks: each turn's thinking block is generated in the output and not carried into later turns' input](https://platform.claude.com/docs/images/context-window-thinking.svg)

1152</Frame>

11481153

11491154The second shows the same regime with tool use: thinking stays in context alongside its tool result for the duration of the assistant turn, then drops out on the next user turn.

11501155

1151![Diagram of thinking with tool use on a model that strips previous thinking blocks: thinking is kept with its tool result, then dropped on the next user turn](https://platform.claude.com/docs/images/context-window-thinking-tools.svg)

1156<Frame>

1157 ![Diagram of thinking with tool use on a model that strips previous thinking blocks: thinking is kept with its tool result, then dropped on the next user turn](https://platform.claude.com/docs/images/context-window-thinking-tools.svg)

1158</Frame>

11521159

11531160Use the [token counting API](https://platform.claude.com/docs/en/build-with-claude/token-counting) to get accurate counts for your specific use case, especially for multi-turn conversations that include thinking.

11541161

from line 1197

11901197

11911198### Sampling parameters

11921199

1193On Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Mythos Preview, Claude Opus 5.5, Claude Opus 5, Claude Opus 4.8, Claude Opus 4.7, Claude Sonnet 5.5, and Claude Sonnet 5, non-default `temperature`, `top_p`, or `top_k` values return a 400 error on every request, regardless of whether thinking is used. On older models, the restriction applies only while thinking is on: `temperature` and `top_k` are incompatible with thinking, and `top_p` is allowed at values between 0.95 and 1.

1200On Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Mythos Preview, Claude Opus 5.5, Claude Opus 5, Claude Opus 4.8, Claude Opus 4.7, Claude Sonnet 5.5, Claude Sonnet 5, and Claude Haiku 5.5, non-default `temperature`, `top_p`, or `top_k` values return a 400 error on every request, regardless of whether thinking is used. On older models, the restriction applies only while thinking is on: `temperature` and `top_k` are incompatible with thinking, and `top_p` is allowed at values between 0.95 and 1.

11941201

11951202### Response prefill and forced tool use

11961203

from line 1224

12171224| Claude Sonnet 5 | 128K | 300K |

12181225| Claude Sonnet 4.6 | 128K | 300K |

12191226| Claude Sonnet 4.5 | 64K | Not available |

1227| Claude Haiku 5.5 | 128K | 300K |

12201228| Claude Haiku 4.5 | 64K | Not available |

12211229

12221230See the [models overview](https://platform.claude.com/docs/en/models/overview) for limits on legacy models.

build-with-claude/thinking-steering-and-cost Changed · +7 / -7 lines

from line 43

4343

4444Effort is the primary steering lever for thinking. Each level sets a different default for how often Claude thinks and how deeply:

4545

46| Effort level | Thinking behavior |

47| ------------------------------------- | ------------------------------------------------------------------------------------------------ |

48| `max` | Claude thinks the most readily and at the greatest depth, with no constraint on thinking length. |

49| `xhigh` | Claude thinks more readily and at greater depth than at `high`, suited to extended exploration. |

50| `high` (default on most models) | Claude thinks on most requests that benefit from it. Provides deep reasoning on complex tasks. |

51| `medium` (default on Claude Opus 5.5) | Claude uses moderate thinking. May skip thinking for simple queries. |

52| `low` | Claude minimizes thinking. Skips thinking for simple tasks where speed matters most. |

46| Effort level | Thinking behavior |

47| ---------------------------------------------------------- | ------------------------------------------------------------------------------------------------ |

48| `max` | Claude thinks the most readily and at the greatest depth, with no constraint on thinking length. |

49| `xhigh` | Claude thinks more readily and at greater depth than at `high`, suited to extended exploration. |

50| `high` (default on most models) | Claude thinks on most requests that benefit from it. Provides deep reasoning on complex tasks. |

51| `medium` (default on Claude Opus 5.5 and Claude Haiku 5.5) | Claude uses moderate thinking. May skip thinking for simple queries. |

52| `low` | Claude minimizes thinking. Skips thinking for simple tasks where speed matters most. |

5353

5454At every level, Claude determines per request whether to think. In a tool-use loop, the first request after new user input typically carries most of the reasoning, and follow-up requests that only process tool results can skip thinking, including at `xhigh` and `max`. Thinking per request also tends to decrease as a conversation grows longer. No level guarantees a thinking block on every request.

5555

build-with-claude/thinking-troubleshooting Changed · +10 / -5 lines

from line 31

from line 39

3940

4041*1 `enabled` and `budget_tokens` still work on these models but are deprecated; use adaptive thinking instead.*\

41*2 Claude Opus 5 accepts `"disabled"` at [effort](https://platform.claude.com/docs/en/build-with-claude/effort) `high` or below; combining it with effort `xhigh` or `max` returns a 400 error. This restriction is enforced on each request.*\

42*2 Claude Opus 5 and Claude Haiku 5.5 accept `"disabled"` at [effort](https://platform.claude.com/docs/en/build-with-claude/effort) `high` or below; combining it with effort `xhigh` or `max` returns a 400 error. On Claude Opus 5, this restriction is enforced on each request. On Claude Haiku 5.5, a per-message effort that differs from the level in effect also returns a 400 error.*\

4243*3 Claude Sonnet 5.5 accepts `"between_tools"` at effort `high` or below. Combining it with effort `xhigh` or `max` returns a 400 error, and so does a per-message effort that differs from the level in effect.*

4344

44Models marked `Always on` cannot turn thinking off. Models marked `On` default to thinking. Claude Opus 5 and Claude Sonnet 5 accept `thinking: {type: "disabled"}`. On Claude Sonnet 5.5, send `thinking: {type: "between_tools"}` to turn off up-front thinking.

45Models marked `Always on` cannot turn thinking off. Models marked `On` default to thinking. Claude Opus 5, Claude Sonnet 5, and Claude Haiku 5.5 accept `thinking: {type: "disabled"}`. On Claude Sonnet 5.5, send `thinking: {type: "between_tools"}` to turn off up-front thinking.

4546

4647Earlier Claude 4 models (Claude Opus 4.1, Claude Sonnet 4, and Claude Opus 4) support extended thinking only. See [Model deprecations](https://platform.claude.com/docs/en/about-claude/model-deprecations) for their availability. Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, and Claude Mythos 5 are not available under [zero data retention](https://platform.claude.com/docs/en/manage-claude/api-and-data-retention#model-specific-data-retention-requirements) unless expressly authorized by Anthropic.

4748

from line 76

7576

7677Omit the `thinking` parameter; these models think without any configuration. If your goal was to keep thinking text out of responses, use `display: "omitted"` instead of disabling thinking; see [Controlling thinking display](https://platform.claude.com/docs/en/build-with-claude/thinking#controlling-thinking-display).

7778

78A 400 error on `"disabled"` can also occur on Claude Opus 5, which accepts `thinking: {type: "disabled"}` only at [effort](https://platform.claude.com/docs/en/build-with-claude/effort) `high` or below: combining it with effort `xhigh` or `max` is rejected. Lower the effort level, or leave thinking on.

79A 400 error on `"disabled"` can also occur on Claude Opus 5 and Claude Haiku 5.5, which accept `thinking: {type: "disabled"}` only at [effort](https://platform.claude.com/docs/en/build-with-claude/effort) `high` or below: combining it with effort `xhigh` or `max` is rejected. Lower the effort level, or set `thinking` to `{"type": "adaptive"}`.

7980

8081On Claude Sonnet 5.5, `thinking: {type: "disabled"}` returns a 400 error at every effort level. The message reads:

8182

from line 110

109110

110111Lower the effort to `high` or below. To run at `xhigh` or `max`, use adaptive thinking: omit the `thinking` field or send `thinking: {"type": "adaptive"}`. That's what the message means by "enable thinking". Claude Sonnet 5.5 rejects `"enabled"` with a 400 error.

111112

113Claude Haiku 5.5 returns the same message when a request sends `thinking: {type: "disabled"}` at effort `xhigh` or `max`. It accepts `"disabled"` only at `high` or below. Lower the effort, or set `thinking` to `{"type": "adaptive"}`.

114

112115## A 400 error says effort cannot change when thinking is disabled

113116

114117On Claude Sonnet 5.5, a request with `thinking: {type: "between_tools"}` whose [per-message effort](https://platform.claude.com/docs/en/build-with-claude/effort#change-effort-mid-conversation-beta) changes the level fails with a 400 error whose message reads:

from line 124

121124

122125Remove that per-message effort, or set it to the level in effect. To vary effort per turn, use adaptive thinking, which is what the message means by "enable thinking".

123126

127Claude Haiku 5.5 returns the same message when a request with `thinking: {type: "disabled"}` sets a per-message effort that differs from the level in effect. Remove that per-message effort, or set `thinking` to `{"type": "adaptive"}` to vary effort per turn.

128

124129## A 400 error says adaptive thinking is not supported

125130

126131The request fails with a 400 error whose message reads:

from line 152

147152

148153## A 400 error says a thinking block signature is invalid

149154

150A request to Claude Fable 5.1, Claude Opus 5.5, or Claude Sonnet 5.5 that replays earlier thinking blocks fails with a 400 `invalid_request_error` whose message reads:

155A request to Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, or Claude Haiku 5.5 that replays earlier thinking blocks fails with a 400 `invalid_request_error` whose message reads:

151156

152157```text wrap

153158messages.{i}.content.{j}: Invalid `signature` in `thinking` block. The block is bound to a different conversation. Remove the block, or set `thinking.block_binding.prefix_mismatch_behavior` to "drop_block".

from line 164

159164

160165If the message stops after ``Invalid `signature` in `thinking` block``, the signature itself didn't verify: it was truncated, altered, or sent back empty, and `prefix_mismatch_behavior` doesn't apply. Edited thinking text returns a different error. See [A 400 error says thinking blocks cannot be modified](https://platform.claude.com/docs/en/build-with-claude/thinking-troubleshooting#error-thinking-blocks-modified).

161166

162On Claude Fable 5.1, Claude Opus 5.5, and Claude Sonnet 5.5, the API accepts a replayed thinking block only while the `system` prompt, `tools`, and messages that preceded it are unchanged. See [Keeping the prefix unchanged](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#prefix-check). The error means something earlier in the conversation changed between requests: an edited, reordered, or removed turn, a per-turn reminder that was injected and later removed, a rebuilt `system` prompt or `tools` array, or client-side compaction that kept recent turns and their thinking verbatim. The check is enforced for new accounts created on or after August 31, 2026, and for any request that sets `thinking.block_binding.prefix_mismatch_behavior`. Server-side [compaction](https://platform.claude.com/docs/en/build-with-claude/compaction) and [context editing](https://platform.claude.com/docs/en/build-with-claude/context-editing) never trigger it.

167On Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, and Claude Haiku 5.5, the API accepts a replayed thinking block only while the `system` prompt, `tools`, and messages that preceded it are unchanged. See [Keeping the prefix unchanged](https://platform.claude.com/docs/en/build-with-claude/preserved-thinking#prefix-check). The error means something earlier in the conversation changed between requests: an edited, reordered, or removed turn, a per-turn reminder that was injected and later removed, a rebuilt `system` prompt or `tools` array, or client-side compaction that kept recent turns and their thinking verbatim. The check is enforced for new accounts created on or after August 31, 2026, and for any request that sets `thinking.block_binding.prefix_mismatch_behavior`. Server-side [compaction](https://platform.claude.com/docs/en/build-with-claude/compaction) and [context editing](https://platform.claude.com/docs/en/build-with-claude/context-editing) never trigger it.

163168

164169To fix it, keep the history append-only: pass earlier turns back exactly as sent and received, add instructions with a [mid-conversation system message](https://platform.claude.com/docs/en/build-with-claude/mid-conversation-system-messages) instead of editing `system` or `tools`, and let server-side [context editing](https://platform.claude.com/docs/en/build-with-claude/context-editing) or [compaction](https://platform.claude.com/docs/en/build-with-claude/compaction) do any trimming. Retrying the same request body doesn't clear the error. To continue this request without the invalidated reasoning, send the `thinking-binding-controls-2026-08-01` beta header and set `thinking.block_binding.prefix_mismatch_behavior` to `"drop_block"`. Alternatively, strip every `thinking` and `redacted_thinking` block from the history (at minimum the named block and every one after it, in that turn and all later turns), leave each turn's other blocks in place, and retry once. On Claude Sonnet 5.5, `block_binding` works only with `thinking: {"type": "adaptive"}`. With `between_tools`, keep the history append-only, or strip the thinking blocks from the edited turn on.

165170

manage-claude/inference-hooks-endpoint Changed · +35 / -17 lines

from line 148

148148

149149Anthropic sends an HTTPS `POST` to the URL your administrator configures. The whole configured URL is the endpoint: there is no fixed path suffix, so choose any path that suits your server.

150150

151Host your AI security server where Anthropic can reach it: an `https://` URL on port 443, on a publicly routable host (private, loopback, and carrier-grade NAT ranges are refused at connect time), with a certificate that validates against the public CA trust store, responding without redirects. The configured URL must be the final destination. Reverse-tunnel hosts (ngrok and similar tunnel services) are not supported: Anthropic's network policy blocks them. Host your server on a domain you control. [Configure Inference hooks](https://platform.claude.com/docs/en/manage-claude/inference-hooks-configuration) covers how your administrator sets and tests the URL.

151Host your AI security server where Anthropic can reach it: an `https://` URL on port 443, on a publicly routable host (private, loopback, and carrier-grade NAT ranges are refused at connect time), with a certificate that validates against the public CA trust store, responding without redirects. The host must have an IPv4 address, which Anthropic uses even when the host also has IPv6 addresses; a URL whose host is `localhost` or an IPv6 address is refused. The configured URL must be the final destination. Reverse-tunnel hosts (ngrok and similar tunnel services) are not supported: Anthropic's network policy blocks them. Host your server on a domain you control. [Configure Inference hooks](https://platform.claude.com/docs/en/manage-claude/inference-hooks-configuration) covers how your administrator sets and tests the URL.

152152

153153Every request carries these fixed headers, along with any [custom request headers](https://platform.claude.com/docs/en/manage-claude/inference-hooks-configuration) your administrator configured and, once your organization has a signing secret, the `webhook-*` signature headers described in [Verify the signature](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#verify-the-signature):

154154

from line 219

219219

220220Each entry in `messages` has a `role` of `user` or `assistant` (tool results appear under the `user` role, matching the public Messages API content model) and a `content` array of blocks discriminated by `type`:

221221

222| Block `type` | Fields |

223| ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

224| `text` | `text`: the text content. |

225| `tool_use` | `id`: the identifier the matching tool result references. `tool_name`: the tool's name. `input`: the arguments the model passed to the tool. `tool_info`: on a tool call frame only (left out elsewhere, never `null`), an object that says who provides the tool; see [The tool call frame](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#the-tool-call-frame). |

226| `tool_result` | `content`: the tool's output as text, with parts joined by newlines; binary parts such as images are replaced by placeholder markers, and raw bytes are never sent. `is_error`: whether the tool call failed. `tool_name`: the tool's name, so a policy can condition on tool identity without cross-referencing an earlier block. `tool_use_id`: the `id` of the matching `tool_use` block. |

227| `attachment` | `file_name`: the original file name or path. `media_type`: the attachment's media type. `size_bytes`: the size of the original file. `text`: the text content of the attachment when available, such as extracted document text, an audio transcript, or link metadata. Raw attachment bytes are never sent. |

222| Block `type` | Fields |

223| ------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

224| `text` | `text`: the text content. |

225| `tool_use` | `id`: the identifier the matching tool result references. `tool_name`: the tool's name. `input`: the arguments the model passed to the tool. `tool_info`: on a tool call frame only (left out elsewhere, never `null`), an object that says who runs or provides the tool; see [The tool call frame](https://platform.claude.com/docs/en/manage-claude/inference-hooks-endpoint#the-tool-call-frame). |

226| `tool_result` | `content`: the tool's output as text, with parts joined by newlines; binary parts such as images are replaced by placeholder markers, and raw bytes are never sent. `is_error`: whether the tool call failed. `tool_name`: the tool's name, so a policy can condition on tool identity without cross-referencing an earlier block. `tool_use_id`: the `id` of the matching `tool_use` block. |

227| `attachment` | `file_name`: the original file name or path. `media_type`: the attachment's media type. `size_bytes`: the size of the original file. `text`: the text content of the attachment when available, such as extracted document text, an audio transcript, or link metadata. Raw attachment bytes are never sent. |

228228

229229Apart from `type`, a `text` block's `text`, and a `tool_result` block's `content` and `is_error`, any of these fields can be `null` when the value isn't known; for example, an image arrives as an `attachment` block with `file_name` and `text` set to `null`.

230230

from line 252

252252

253253* `type` is `"tool_call"`.

254254* `messages` holds only the latest message, the `assistant` message Claude just produced: any `text` blocks and one `tool_use` block per tool call the frame lists, in the order the model produced them. Earlier conversation is left out, because the prompt frame sent before that model call carried it. Read the last entry of `messages`, because the protocol may later add earlier messages before it.

255* Each `tool_use` block carries a `tool_info` object that says who provides the tool.

255* Each `tool_use` block carries a `tool_info` object that says who runs or provides the tool.

256256

257257Where `session_id` is set, it is the same on both frames. The tool call frame has its own `request_id`, which is opaque like the prompt frame's.

258258

259`tool_info` says who provides the tool, not who runs it or what it can reach. It is one of four kinds, told apart by its `type` field, and each kind carries its own fields. Anthropic sends the first of the following kinds that fits the tool. New kinds may appear: accept a `type` you don't recognize, and for such a kind rely only on `type`.

259`tool_info` says who runs or provides the tool, not what the tool can reach. It is one of four kinds, told apart by its `type` field, and each kind carries its own fields. Anthropic sends the first of the following kinds that fits the tool. New kinds may appear: accept a `type` you don't recognize, and for such a kind rely only on `type`.

260260

261261An optional field that doesn't apply is left out, never `null`, so a `tool_info` can be just `{"type": "client"}`. `tool_name` is chosen by whoever defined the tool, and a server's `toolset_name` by whoever wrote the request, so don't treat either as a trust boundary.

262262

263263### Platform tools

264264

265A platform tool is one of the Claude API's predefined tools, such as web search or bash. It is `"platform"` whether Anthropic or the application that calls Claude runs it.

265A platform tool is one that the Claude API itself runs while it serves the request, such as web search or code execution.

266266

267267| Field | Present | Description |

268268| -------------- | -------- | ------------------------------------------------------------------------------------------------------------------- |

269269| `type` | Always | `"platform"` |

270270| `tool_type` | Always | The tool's versioned type, such as `web_search_20250305`. Match it exactly; don't parse a name or a date out of it. |

271| `toolset_name` | Optional | The toolset the tool belongs to, such as `browser`. |

271| `toolset_name` | Optional | The group of tools it belongs to. |

272272

273273### Application tools

274274

275An application tool is one that the Anthropic application making the request provides itself, such as claude.ai's own tools. It is `"application"` whether Anthropic or the application that calls Claude runs it.

275An application tool is one that the Anthropic application making the request provides itself, such as claude.ai's own tools.

276276

277277| Field | Present | Description |

278278| -------------- | -------- | --------------------------------- |

from line 311

311311

312312### Client tools

313313

314A client tool is any other tool. The application that calls Claude declares it and receives its calls. A call to a tool the request doesn't declare is also `"client"`.

314A client tool is any other tool, usually one that the application calling Claude runs. A tool that Anthropic defines but the calling application runs, such as bash or computer use, is a client tool too, and carries its versioned type in `tool_type`. A call to a tool the request doesn't declare is also `"client"`.

315315

316316A tool on an MCP server that Claude Code connects to directly from the user's computer, such as a local MCP server, is a client tool. Its `tool_info` is `{"type": "client"}`, and its `tool_name` is the name Claude Code gives it, in the form `mcp__<server>__<tool>`.

317317

318318A deny stops the call before Claude Code runs it. The exchange between Claude Code and a local server doesn't pass through Anthropic, so your AI security server sees the tool's result only when Claude Code sends it back. Its text is then in a `tool_result` block in the next prompt frame.

319319

320| Field | Present | Description |

321| -------------- | -------- | --------------------------------- |

322| `type` | Always | `"client"` |

323| `toolset_name` | Optional | The group of tools it belongs to. |

320| Field | Present | Description |

321| -------------- | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------- |

322| `type` | Always | `"client"` |

323| `tool_type` | Optional | The tool's versioned type, such as `bash_20250124`, when Anthropic defines the tool. Match it exactly; don't parse a name or a date out of it. |

324| `toolset_name` | Optional | The group of tools it belongs to, such as `browser`. |

324325

325326A `tool_use` block for a tool that the calling application declares:

326327

from line 335

334335 },

335336 "tool_info": {

336337 "type": "client"

338 }

339}

340```

341

342A `tool_use` block for bash, which Anthropic defines and the calling application runs:

343

344```json

345{

346 "type": "tool_use",

347 "id": "toolu_01HiJkLmNoPqRsTuVwXyZaBc",

348 "tool_name": "bash",

349 "input": {

350 "command": "ls -la reports/"

351 },

352 "tool_info": {

353 "type": "client",

354 "tool_type": "bash_20250124"

337355 }

338356}

339357```

managed-agents/reference Changed · +3 / -3 lines

from line 76

7676 </Tab>

7777

7878 <Tab title="System events">

79 | Type | Description |

80 | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |

81 | `system.message` | Append privileged system-level context that applies to the accompanying turn and all subsequent turns. Supported on Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Opus 5.5, Claude Opus 5, Claude Opus 4.8, and Claude Sonnet 5.5. On an unsupported primary model the event is rejected with `model_does_not_support_mid_conversation_system`. |

79 | Type | Description |

80 | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |

81 | `system.message` | Append privileged system-level context that applies to the accompanying turn and all subsequent turns. Supported on Claude Fable 5.1, Claude Mythos 5.1, Claude Fable 5, Claude Mythos 5, Claude Opus 5.5, Claude Opus 5, Claude Opus 4.8, Claude Sonnet 5.5, and Claude Haiku 5.5. On an unsupported primary model the event is rejected with `model_does_not_support_mid_conversation_system`. |

8282 </Tab>

8383

8484 <Tab title="Event deltas">

models/fable-5-1/overview Changed · +7 / -7 lines

from line 28

2828

2929## How it compares

3030

32| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------- | :------------------- | :------------- | :--------------- |

33| **Claude Fable 5.1** (this model) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

34| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Moderate | Adaptive (always on) | `medium` | Jun 2026 |

35| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Fast | Adaptive | `high` | Jun 2026 |

36| [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview) | 200K | 64K | $1 / $5 | Fastest | Extended | — | Feb 2025 |

32| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :------- | :------------------- | :------------- | :--------------- |

33| **Claude Fable 5.1** (this model) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

34| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Moderate | Adaptive (always on) | `medium` | Jun 2026 |

35| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Fast | Adaptive | `high` | Jun 2026 |

36| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Fastest | Adaptive | `medium` | Jun 2026 |

3737

3838* **Context:** 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.

39* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

39* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

4040* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

4141* **Latency:** Comparative latency, relative to the current lineup, as published in the models overview. Actual latency depends on prompt length, output length, and thinking effort.

4242* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

models/fable-5/overview Changed · +8 / -8 lines

from line 20

2020

2121## How it compares to the current lineup

2222

24| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------------------- | :------------- | :--------------- |

25| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

26| **Claude Fable 5** (this model) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jan 2026 |

27| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

28| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

29| [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview) | 200K | 64K | $1 / $5 | Extended | — | Feb 2025 |

24| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :------------------- | :------------- | :--------------- |

25| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

26| **Claude Fable 5** (this model) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jan 2026 |

27| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

28| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

29| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Adaptive | `medium` | Jun 2026 |

3030

3131* **Context:** 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.

32* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

32* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

3333* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

3434* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

3535* **Default effort:** The effort parameter’s default on the Claude API. Models without a value don’t support the parameter.

models/haiku-4-5/overview Changed · +23 / -14 lines

from line 1

11---

22title: Claude Haiku 4.5

33url: https://platform.claude.com/docs/en/models/haiku-4-5/overview

4description: "Claude Haiku 4.5 at a glance: what it's for, model IDs on every platform, context window, output limits, pricing, availability, and the guides and resources for building with it."

4description: "Claude Haiku 4.5 reference: lifecycle status, model IDs on every platform, context window, output limits, pricing, and migration resources. Claude Haiku 5.5 is the current Haiku model."

55---

66

7**Latest.** Released October 15, 2025.

7**Legacy.** Released October 15, 2025.

88

99The fastest model with near-frontier intelligence

1010

11Although Claude Haiku 4.5 is still available, you should consider migrating to Claude Haiku 5.5 for improved performance. [See Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) · [Migrate to Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/migration-guide)

12

1113Model ID: `claude-haiku-4-5-20251001`

1214

1315Context window: 200K tokens · Max output: 64K tokens · Input pricing: $1 / MTok · Output pricing: $5 / MTok

1416

15[Announcement](https://www.anthropic.com/news/claude-haiku-4-5) · [Migration guide](https://platform.claude.com/docs/en/models/haiku-4-5/migration-guide)

17[Announcement](https://www.anthropic.com/news/claude-haiku-4-5)

1618

1719## How it compares

1820

20| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------- | :------------------- | :------------- | :--------------- |

21| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

22| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Moderate | Adaptive (always on) | `medium` | Jun 2026 |

23| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Fast | Adaptive | `high` | Jun 2026 |

24| **Claude Haiku 4.5** (this model) | 200K | 64K | $1 / $5 | Fastest | Extended | — | Feb 2025 |

22| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :------------------- | :------------- | :--------------- |

23| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

24| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

25| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

26| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Adaptive | `medium` | Jun 2026 |

27| **Claude Haiku 4.5** (this model) | 200K | 64K | $1 / $5 | Extended | — | Feb 2025 |

2528

2629* **Context:** 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.

27* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

30* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

2831* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

29* **Latency:** Comparative latency, relative to the current lineup, as published in the models overview. Actual latency depends on prompt length, output length, and thinking effort.

3032* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

3133* **Default effort:** The effort parameter’s default on the Claude API. Models without a value don’t support the parameter.

3234* **Knowledge cutoff:** Reliable knowledge cutoff: the date through which the model’s knowledge is most extensive and reliable.

from line 68

6668| Max output | 64K tokens |

6769| [Thinking](https://platform.claude.com/docs/en/build-with-claude/thinking) | Extended |

6870| [Default effort](https://platform.claude.com/docs/en/build-with-claude/effort) | Not supported |

69| Comparative latency | Fastest |

7071| Input → output | Text and images → text |

7172| Reliable knowledge cutoff | Feb 2025 |

7273| Training data cutoff | Jul 2025 |

from line 76

7576

7677| Feature | Value |

7778| :---------------------------------------------------------------------------- | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

78| [Status](https://platform.claude.com/docs/en/about-claude/model-deprecations) | Active (latest) |

79| [Status](https://platform.claude.com/docs/en/about-claude/model-deprecations) | Active (legacy) |

7980| Released | October 15, 2025 |

8081| Retirement | Not sooner than October 15, 2026 |

8182| Platforms | Claude API, [Amazon Bedrock](https://platform.claude.com/docs/en/build-with-claude/claude-in-amazon-bedrock), [Amazon Bedrock (InvokeModel)](https://platform.claude.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy), [Google Cloud](https://platform.claude.com/docs/en/build-with-claude/claude-on-vertex-ai), [Microsoft Foundry](https://platform.claude.com/docs/en/build-with-claude/claude-in-microsoft-foundry), [Claude Platform on AWS](https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws) |

from line 90

8990## Resources

9091

9192<CardGroup cols={3}>

93 <Card title="Migrate to Claude Haiku 5.5" icon="arrows-left-right" href="https://platform.claude.com/docs/en/models/haiku-5-5/migration-guide">

94 What changes when moving from Claude Haiku 4.5 to Claude Haiku 5.5.

95 </Card>

96

97 <Card title="Claude Haiku 5.5" icon="arrow-right" href="https://platform.claude.com/docs/en/models/haiku-5-5/overview">

98 The current Haiku model: overview, specs, and resources.

99 </Card>

100

92101 <Card title="Extended thinking" icon="brain" href="https://platform.claude.com/docs/en/build-with-claude/extended-thinking">

93102 Claude Haiku 4.5 supports manual extended thinking with `budget_tokens`.

94103 </Card>

from line 107

98107 </Card>

99108

100109 <Card title="Reduce latency" icon="gauge" href="https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-latency">

101 Techniques that pair well with the fastest model in the lineup.

110 Techniques that pair well with a fast, low-cost model.

102111 </Card>

103112</CardGroup>

104113

models/mythos-5-1/overview Changed · +8 / -8 lines

from line 18

1818

1919## How it compares

2020

22| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------- | :------------------- | :------------- | :--------------- |

23| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

24| **Claude Mythos 5.1** (this model) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

25| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Moderate | Adaptive (always on) | `medium` | Jun 2026 |

26| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Fast | Adaptive | `high` | Jun 2026 |

27| [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview) | 200K | 64K | $1 / $5 | Fastest | Extended | — | Feb 2025 |

22| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :------- | :------------------- | :------------- | :--------------- |

23| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

24| **Claude Mythos 5.1** (this model) | 1M | 128K | $10 / $50 | Slower | Adaptive (always on) | `high` | Jun 2026 |

25| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Moderate | Adaptive (always on) | `medium` | Jun 2026 |

26| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Fast | Adaptive | `high` | Jun 2026 |

27| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Fastest | Adaptive | `medium` | Jun 2026 |

2828

2929* **Context:** 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.

30* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

30* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

3131* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

3232* **Latency:** Comparative latency, relative to the current lineup, as published in the models overview. Actual latency depends on prompt length, output length, and thinking effort.

3333* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

models/mythos-5/overview Changed · +8 / -8 lines

from line 18

1818

1919## How it compares

2020

22| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------------------- | :------------- | :--------------- |

23| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

24| **Claude Mythos 5** (this model) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jan 2026 |

25| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

26| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

27| [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview) | 200K | 64K | $1 / $5 | Extended | — | Feb 2025 |

22| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :------------------- | :------------- | :--------------- |

23| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

24| **Claude Mythos 5** (this model) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jan 2026 |

25| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

26| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

27| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Adaptive | `medium` | Jun 2026 |

2828

2929* **Context:** 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.

30* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

30* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

3131* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

3232* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

3333* **Default effort:** The effort parameter’s default on the Claude API. Models without a value don’t support the parameter.

models/opus-4-5/overview Changed · +8 / -8 lines

from line 16

1616

1717## How it compares to the current lineup

1818

20| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------------------- | :------------- | :--------------- |

21| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

22| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

23| **Claude Opus 4.5** (this model) | 200K | 64K | $5 / $25 | Extended | `high` | May 2025 |

24| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

25| [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview) | 200K | 64K | $1 / $5 | Extended | — | Feb 2025 |

20| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :------------------- | :------------- | :--------------- |

21| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

22| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

23| **Claude Opus 4.5** (this model) | 200K | 64K | $5 / $25 | Extended | `high` | May 2025 |

24| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

25| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Adaptive | `medium` | Jun 2026 |

2626

2727* **Context:** 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.

28* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

28* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

2929* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

3030* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

3131* **Default effort:** The effort parameter’s default on the Claude API. Models without a value don’t support the parameter.

models/opus-4-6/overview Changed · +8 / -8 lines

from line 16

1616

1717## How it compares to the current lineup

1818

20| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :----------------------------- | :------------- | :--------------- |

21| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

22| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

23| **Claude Opus 4.6** (this model) | 1M | 128K | $5 / $25 | Adaptive (extended deprecated) | `high` | May 2025 |

24| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

25| [Claude Haiku 4.5](https://platform.claude.com/docs/en/models/haiku-4-5/overview) | 200K | 64K | $1 / $5 | Extended | — | Feb 2025 |

20| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------------- | :----------------------------- | :------------- | :--------------- |

21| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

22| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

23| **Claude Opus 4.6** (this model) | 1M | 128K | $5 / $25 | Adaptive (extended deprecated) | `high` | May 2025 |

24| [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | 1M | 128K | $2 / $10 | Adaptive | `high` | Jun 2026 |

25| [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | 1M | 128K | From $0.10 / $0.50 | Adaptive | `medium` | Jun 2026 |

2626

2727* **Context:** 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.

28* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, 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.

28* **Max output:** Synchronous Messages API limit. On the Message Batches API, Claude Opus 5.5, Claude Opus 5, Claude Sonnet 5.5, Claude Sonnet 5, Claude Haiku 5.5, 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.

2929* **Price / MTok:** Input / output, base price per million tokens. Batch API requests are 50% off; prompt caching reads cost 10% of the base input price (2.5% on Claude Fable 5.1 and Claude Mythos 5.1, 5% on Claude Opus 5.5). See Pricing for the full list.

3030* **Thinking:** Adaptive thinking lets the model decide how much to think, steered by effort. Extended thinking is the manual budget\_tokens mode on earlier models.

3131* **Default effort:** The effort parameter’s default on the Claude API. Models without a value don’t support the parameter.

models/opus-4-7/overview Changed · +8 / -8 lines

from line 16

1616

1717## How it compares to the current lineup

1818

20| :---------------------------------------------------------------------------------- | :------ | :--------- | :----------- | :------------------- | :------------- | :--------------- |

21| [Claude Fable 5.1](https://platform.claude.com/docs/en/models/fable-5-1/overview) | 1M | 128K | $10 / $50 | Adaptive (always on) | `high` | Jun 2026 |

22| [Claude Opus 5.5](https://platform.claude.com/docs/en/models/opus-5-5/overview) | 1M | 128K | $4 / $20 | Adaptive (always on) | `medium` | Jun 2026 |

23| **Claude Opus 4.7** (this model) | 1M | 128K | $5 / $25 | Adaptive | `high` | Jan 2026 |