Follow Discord
Sweep 02 Oct 2026 · 18:55Z Build v2.1.288 509 read Stable v2.1.285 Latest v2.1.288 Next v2.1.288 Feeds RSS JSON llms.txt llms-full.txt Unofficial
HomeDocsClaude Developer Platform 1 Oct 2026
One change · api

refusals-and-fallback changedbuild-with-claude/refusals-and-fallback

Nearest release: v2.1.287, published 4 hours before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.

Recorded here
Lines+22added
Lines−9removed
From line 180 where the diff opens
First seen 14 Aug 2026 this site's first read of the page
Recorded edits15to this page, all time

### Keep reasoning in thinking blocks

The whole hunk

from line 180, old and new numbered
/
lines
from line 180
180180* `category` and `explanation` are both `null` when the refusal does not map to a named category. That `null` is a normal, permanent value, not a placeholder.
181181* `stop_details` itself is `null` for every stop reason other than `refusal`.
182182 
183| `category` | What it means | Billed before any output |
184| ------------------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------ |
185| `"cyber"` | The request could enable cyber harm, such as malware or exploit development. Benign cybersecurity work can also trigger this category. | No |
186| `"bio"` | The request could enable biological harm, such as dangerous lab methods. Beneficial life sciences work can also trigger this category. | Yes |
187| `"frontier_llm"` | The request could assist the development of competing AI models, which is restricted under [Anthropic's commercial terms](https://www.anthropic.com/legal/commercial-terms). Benign machine learning work can also trigger this category. | Yes |
188| `"reasoning_extraction"` | The request asks the model to reproduce its internal reasoning in the response text. To get reasoning in a structured form instead, use [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking). | Yes |
189| `"general_harms"` | The request falls under a usage-policy area outside the four named categories. Benign work can also trigger this category. | No |
183| `category` | What it means | Billed before any output |
184| ------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ------------------------ |
185| `"cyber"` | The request could enable cyber harm, such as malware or exploit development. Benign cybersecurity work can also trigger this category. | No |
186| `"bio"` | The request could enable biological harm, such as dangerous lab methods. Beneficial life sciences work can also trigger this category. | Yes |
187| `"frontier_llm"` | The request could assist the development of competing AI models, which is restricted under [Anthropic's commercial terms](https://www.anthropic.com/legal/commercial-terms). Benign machine learning work can also trigger this category. | Yes |
188| `"reasoning_extraction"` | The request asks the model to reproduce its internal reasoning in the response text. To get reasoning in a structured form instead, use [adaptive thinking](https://platform.claude.com/docs/en/build-with-claude/thinking). See [Keep reasoning in thinking blocks](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#keep-reasoning-in-thinking-blocks). | Yes |
189| `"general_harms"` | The request falls under a usage-policy area outside the four named categories. Benign work can also trigger this category. | No |
190190 
191191A refusal can arrive before any output, or mid-stream after partial output. In either case, treat any partial output as incomplete and discard it.
192192 
193### Keep reasoning in thinking blocks
194 
195A `reasoning_extraction` refusal usually comes from a prompt that asks the model to put its thinking or reasoning in the output, either verbatim or in a fixed format. Common examples:
196 
197* A `<thinking>`, `<reasoning>`, or scratchpad section that the model fills in before it answers
198* A `reasoning`, `thinking`, or `trace` field in JSON output or in a tool input
199* The model's private notes, or a running log of its reasoning
200* Reasoning asked for verbatim or in full
201 
202The wording can sit in a system prompt, a skill, or a tool description, so check those too.
203 
204You can still ask Claude to explain its answer. Ask for a short explanation, the evidence behind a result, or a summary of the actions it took. To see how the model reasoned, set `display: "summarized"` and read the [`thinking` blocks](https://platform.claude.com/docs/en/build-with-claude/thinking#summarized-thinking). This category has no recommended fallback model, so change the prompt rather than retry the request. If a reworded request is still declined, [contact support](https://support.claude.com/en/articles/9015913-how-to-get-support) and include the [request ID](https://platform.claude.com/docs/en/api/errors#request-id).
205 
193206## How refusals are billed
194207 
195208These billing rules apply on every platform: the Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, and Microsoft Foundry.
from line 1128
11151128 
11161129* Retries walk your fallback list in order. A fallback model that itself refuses passes the request to the next entry.
11171130* 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.
1118* Thinking blocks from Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 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)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API.
1131* Thinking blocks from Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 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)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API and Google Cloud.
11191132* 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.
11201133* 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.
11211134 
from line 1148
11351148 <Step title="Re-send on a fallback model">
11361149 Send the same request with `model` set to a fallback model, such as Claude Opus 4.8. If the refused request sent `thinking: {"type": "between_tools"}`, change `thinking` first: only Claude Sonnet 5.5 accepts that value, so omit `thinking` or set a value the fallback model accepts. [Server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback) under the `2026-07-01` header makes this change for you when it falls back to Claude Sonnet 5. 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):
11371150 
1138 * **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)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API.
1151 * **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)). It drops Claude Sonnet 5.5 blocks too, for every fallback model except Claude Opus 5.5 on the Claude API and Google Cloud.
11391152 * **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)).
11401153 </Step>
11411154 
Feedback