Follow Discord
Sweep 23 Oct 2026 · 21:48Z Build v2.1.310 529 read Stable v2.1.308 Latest v2.1.310 Next v2.1.310 Feeds RSS JSON llms.txt llms-full.txt Unofficial
One change · api

Beta changedapi/beta

Nearest release: v2.1.296, 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+5,834added
Lines−2,504removed
From line 1,544 where the diff opens
First seen 14 Aug 2026 this site's first read of the page
Recorded edits40to this page, all time

The whole hunk

from line 1544, old and new numbered
/
lines

This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.

from line 1544
15441544 
15451545 Each input message must be an object with a `role` and `content`. You can specify a single `user`-role message, or you can include multiple `user` and `assistant` messages.
15461546 
1547 If the final message uses the `assistant` role, the response content will continue immediately from the content in that message. This can be used to constrain part of the model's response.
1547 If the final message uses the `assistant` role, the response content will continue immediately from the content in that message. This can be used to constrain part of the model's response. This is called prefill. On models that don't support prefill, creating a message that ends with a partial `assistant` response returns a 400 error. See [Prefill not supported](https://platform.claude.com/docs/en/api/errors#prefill-not-supported).
15481548 
15491549 Example with a single `user` message:
15501550 
from line 1562
15621562 ]
15631563 ```
15641564 
1565 Example with a partially-filled response from Claude:
1565 Example with a partially-filled response from Claude, for models that support prefill:
15661566 
15671567 ```json
15681568 [
from line 4549
45494549 
45504550 - `"claude-haiku-4-5"`
45514551 
4552 Fastest model with near-frontier intelligence
4553 
45544552 - `"claude-haiku-4-5-20251001"`
45554553 
4556 Fastest model with near-frontier intelligence
4557 
45584554 - `"claude-opus-4-5"`
45594555 
45604556 Powerful intelligence for long-running agents and coding
from line 5145
51495145 
51505146- `fallbacks: optional BetaFallbacksParam or null`
51515147 
5152 Opt-in server-side retry on one or more substitute models when the requested model declines for policy reasons. Tried in order: if the first entry also declines, the second is tried, and so on. The string "default" requests the requested model's server-defined default fallback configuration.
5148 Opt-in server-side retry on one or more substitute models when the requested model declines for policy reasons. Tried in order: if the first entry also declines, the second is tried, and so on. Some models don't support fallbacks; on those models, a list of fallback models returns a 400 error. See [Server-side fallback](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#server-side-fallback). The string "default" requests the requested model's server-defined default fallback configuration. On a model that doesn't support fallbacks, the request runs on the requested model alone, so a declined request stays declined.
51535149 
51545150 - `array of BetaFallbackParam`
51555151 
from line 5243
52475243 
52485244 - `display: optional "summarized" or "omitted" or "updates" or null`
52495245 
5250 Controls how thinking content appears in the response. When set to `summarized`, thinking is returned normally. When set to `omitted`, thinking content is redacted but a signature is returned for multi-turn continuity. Defaults to `summarized`.
5246 Controls how thinking content appears in the response. When set to `summarized`, thinking is returned normally. When set to `omitted`, thinking content is redacted but a signature is returned for multi-turn continuity. The default depends on the model; see [Controlling thinking display](https://platform.claude.com/docs/en/build-with-claude/thinking#controlling-thinking-display).
52515247 
52525248 - `"summarized"`
52535249 
from line 5269
52735269 
52745270 - `display: optional "summarized" or "omitted" or "updates" or null`
52755271 
5276 Controls how thinking content appears in the response. When set to `summarized`, thinking is returned normally. When set to `omitted`, thinking content is redacted but a signature is returned for multi-turn continuity. Defaults to `summarized`.
5272 Controls how thinking content appears in the response. When set to `summarized`, thinking is returned normally. When set to `omitted`, thinking content is redacted but a signature is returned for multi-turn continuity. The default depends on the model; see [Controlling thinking display](https://platform.claude.com/docs/en/build-with-claude/thinking#controlling-thinking-display).
52775273 
52785274 - `"summarized"`
52795275 
from line 5375
53795375 
53805376- `thinking: optional BetaThinkingConfigParam`
53815377 
5382 Configuration for enabling Claude's extended thinking.
5378 Configuration for Claude's thinking.
53835379 
5384 When enabled, responses include `thinking` content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your `max_tokens` limit.
5380 With `{"type": "adaptive"}`, Claude decides when and how much to think. With `{"type": "enabled"}` (manual extended thinking), you set a `budget_tokens` of at least 1,024. Thinking tokens count toward your `max_tokens` limit.
53855381 
5386 See [extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) for details.
5382 Which `type` values are accepted, and what happens when you omit `thinking`, depend on the model. See [thinking](https://platform.claude.com/docs/en/build-with-claude/thinking#configuring-thinking) for each model's behavior.
53875383 
53885384 - `BetaThinkingConfigEnabled object`
53895385 
from line 7725
77297725 
77307726 Name of the tool.
77317727 
7732 This is how the tool will be called by the model and in `tool_use` blocks.
7733 
7734 - `allowed_callers: optional array of "direct" or "code_execution_20250825" or "code_execution_20260120" or "code_execution_20260521"`
7735 
7736 - `"direct"`
7737 
7738 - `"code_execution_20250825"`
7739 
7740 - `"code_execution_20260120"`
7741 
7742 - `"code_execution_20260521"`
7743 
7744 - `cache_control: optional BetaCacheControlEphemeral or null`
7745 
7746 Create a cache control breakpoint at this content block.
7747 
7748 - `defer_loading: optional boolean`
7749 
7750 If true, tool will not be included in initial system prompt. Only loaded when returned via tool_reference from tool search.
7751 
7752 - `input_examples: optional array of map[unknown]`
7753 
7754 - `strict: optional boolean`
7755 
7756 When true,
7728 This is how the tool will be called by
Feedback