Follow Discord
Sweep 09 Oct 2026 · 17:27Z Build v2.1.296 517 read Stable v2.1.287 Latest v2.1.296 Next v2.1.296 Feeds RSS JSON llms.txt llms-full.txt Unofficial
One change · api

Messages changedapi/beta/messages

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+35added
Lines−182removed
From line 161 where the diff opens
First seen 14 Aug 2026 this site's first read of the page
Recorded edits23to this page, all time

The whole hunk

from line 161, 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 161
161161 
162162 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.
163163 
164 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.
164 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).
165165 
166166 Example with a single `user` message:
167167 
from line 179
179179 ]
180180 ```
181181 
182 Example with a partially-filled response from Claude:
182 Example with a partially-filled response from Claude, for models that support prefill:
183183 
184184 ```json
185185 [
from line 3166
31663166 
31673167 - `"claude-haiku-4-5"`
31683168 
3169 Fastest model with near-frontier intelligence
3170 
31713169 - `"claude-haiku-4-5-20251001"`
31723170 
3173 Fastest model with near-frontier intelligence
3174 
31753171 - `"claude-opus-4-5"`
31763172 
31773173 Powerful intelligence for long-running agents and coding
from line 3762
37663762 
37673763- `fallbacks: optional BetaFallbacksParam or null`
37683764 
3769 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.
3765 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.
37703766 
37713767 - `array of BetaFallbackParam`
37723768 
from line 3860
38643860 
38653861 - `display: optional "summarized" or "omitted" or "updates" or null`
38663862 
3867 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`.
3863 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).
38683864 
38693865 - `"summarized"`
38703866 
from line 3886
38903886 
38913887 - `display: optional "summarized" or "omitted" or "updates" or null`
38923888 
3893 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`.
3889 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).
38943890 
38953891 - `"summarized"`
38963892 
from line 3992
39963992 
39973993- `thinking: optional BetaThinkingConfigParam`
39983994 
3999 Configuration for enabling Claude's extended thinking.
3995 Configuration for Claude's thinking.
40003996 
4001 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.
3997 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.
40023998 
4003 See [extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) for details.
3999 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.
40044000 
40054001 - `BetaThinkingConfigEnabled object`
40064002 
from line 7032
70367032 
70377033 - `"claude-haiku-4-5"`
70387034 
7039 Fastest model with near-frontier intelligence
7040 
70417035 - `"claude-haiku-4-5-20251001"`
70427036 
7043 Fastest model with near-frontier intelligence
7044 
70457037 - `"claude-opus-4-5"`
70467038 
70477039 Powerful intelligence for long-running agents and coding
from line 7174
71827174 
71837175 If true, tool will not be included in initial system prompt. Only loaded when returned via tool_reference from tool search.
71847176 
7185 - `strict: optional boolean`
7186 
7187 When true, guarantees schema validation on tool names and inputs
7188 
7189 - `BetaMCPToolset object`
7190 
7191 Configuration for a group of tools from an MCP server.
7192 
7193 Allows configuring enabled status and defer_loading for all tools
7194 from an MCP server, with optional per-tool overrides.
7195 
7196 - `type: "mcp_toolset"`
7197 
7198 - `mcp_server_name: string`
7199 
7200 Name of the MCP server to configure tools for
7201 
7202 minLength: 1, maxLength: 255
7203 
7204 - `cache_control: optional BetaCacheControlEphemeral or null`
7205 
7206 Create a cache control breakpoint at this content block.
7207 
7208
7177 - `strict
Feedback