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

Create a Message changedapi/beta/messages/create

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+10added
Lines−20removed
From line 159 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 159, 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 159
159159 
160160 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.
161161 
162 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.
162 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).
163163 
164164 Example with a single `user` message:
165165 
from line 177
177177 ]
178178 ```
179179 
180 Example with a partially-filled response from Claude:
180 Example with a partially-filled response from Claude, for models that support prefill:
181181 
182182 ```json
183183 [
from line 3164
31643164 
31653165 - `"claude-haiku-4-5"`
31663166 
3167 Fastest model with near-frontier intelligence
3168 
31693167 - `"claude-haiku-4-5-20251001"`
31703168 
3171 Fastest model with near-frontier intelligence
3172 
31733169 - `"claude-opus-4-5"`
31743170 
31753171 Powerful intelligence for long-running agents and coding
from line 3760
37643760 
37653761- `fallbacks: optional BetaFallbacksParam or null`
37663762 
3767 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.
3763 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.
37683764 
37693765 - `array of BetaFallbackParam`
37703766 
from line 3858
38623858 
38633859 - `display: optional "summarized" or "omitted" or "updates" or null`
38643860 
3865 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`.
3861 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).
38663862 
38673863 - `"summarized"`
38683864 
from line 3884
38883884 
38893885 - `display: optional "summarized" or "omitted" or "updates" or null`
38903886 
3891 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`.
3887 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).
38923888 
38933889 - `"summarized"`
38943890 
from line 3990
39943990 
39953991- `thinking: optional BetaThinkingConfigParam`
39963992 
3997 Configuration for enabling Claude's extended thinking.
3993 Configuration for Claude's thinking.
39983994 
3999 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.
3995 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.
40003996 
4001 See [extended thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking) for details.
3997 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.
40023998 
40033999 - `BetaThinkingConfigEnabled object`
40044000 
from line 7030
70347030 
70357031 - `"claude-haiku-4-5"`
70367032 
7037 Fastest model with near-frontier intelligence
7038 
70397033 - `"claude-haiku-4-5-20251001"`
70407034 
7041 Fastest model with near-frontier intelligence
7042 
70437035 - `"claude-opus-4-5"`
70447036 
70457037 Powerful intelligence for long-running agents and coding
from line 7172
71807172 
71817173 If true, tool will not be included in initial system prompt. Only loaded when returned via tool_reference from tool search.
71827174 
7183 - `strict: optional boolean`
7184 
7185 When true, guarantees schema validation on tool names and inputs
7186 
7187 - `BetaMCPToolset object`
7188 
7189 Configuration for a group of tools from an MCP server.
7190 
7191 Allows configuring enabled status and defer_loading for all tools
7192 from an MCP server, with optional per-tool overrides.
7193 
7194 - `type: "mcp_toolset"`
7195 
7196 - `mcp_server_name: string`
7197 
7198 Name of the MCP server to configure tools for
7199 
7200 minLength: 1, maxLength: 255
7201 
7202 - `cache_control: optional BetaCacheControlEphemeral or null`
7203 
7204 Create a cache control breakpoint at this content block.
7205 
7206
7175 - `strict:
Feedback