> ## Documentation Index
> Fetch the complete documentation index at: https://apidoc.cometapi.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Opprett en melding
> Kall Claude via CometAPI Messages-endepunktet med tekst- og bildeinndata, adaptiv tenkning, Prompt-caching, Streaming og verktøy.
Bruk `POST /v1/messages` til å sende Claude-forespørsler i Anthropic Messages-formatet.
Eksemplene konfigurerer den offisielle Anthropic SDK-en med CometAPI-base-URL-en og
leser API-nøkkelen din fra `$COMETAPI_KEY`.
For feltdefinisjoner og modellspesifikke alternativer kan du se den offisielle
[Messages API-referansen](https://platform.claude.com/docs/en/api/messages/create).
For OpenAI-kompatible forespørsler, se [Chat Completions](/api/text/chat).
Autentiser med `x-api-key` eller `Authorization: Bearer`. Anthropic SDK-en bruker
`x-api-key`. HTTP-eksemplene inkluderer `anthropic-version: 2023-06-01`.
## Hurtigstart
Eksemplene nedenfor ber om tre søkespørringer. Angi `$COMETAPI_KEY` før
du kjører dem. Installer `anthropic` for Python eller `@anthropic-ai/sdk` for JavaScript:
```bash Shell theme={null}
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "Suggest three search queries about renewable energy storage."
}
]
}'
```
```python Python theme={null}
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
message = client.messages.create(
model="claude-opus-5",
max_tokens=1000,
messages=[
{
"role": "user",
"content": "Suggest three search queries about renewable energy storage.",
},
],
)
for block in message.content:
if block.type == "text":
print(block.text)
```
```javascript JavaScript theme={null}
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const message = await client.messages.create({
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "Suggest three search queries about renewable energy storage."
}
]
});
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
```
Svaret inneholder en `content`-array. Les blokker der `type` er `text`;
blokker for tenkning og verktøy kan vises før teksten.
## Kontroller adaptiv tenkning
Sett `thinking.type` til `adaptive`, og velg en `output_config.effort`-verdi.
Følgende eksempel bruker `xhigh` og leser den fullførte meldingen fra en strøm:
```python theme={null}
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=4096,
messages=[
{
"role": "user",
"content": "A shop sells notebooks for $7 each. Buying three earns a $4 discount. A customer pays $20 for three notebooks. Calculate the total and the change, and explain your steps.",
},
],
thinking={
"type": "adaptive",
},
output_config={
"effort": "xhigh",
},
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
```
Tenkning bidrar til utdata-grensen for `max_tokens`. La det være plass til det endelige
svaret samt tenkning. Et svar kan inneholde tekst uten en synlig
tenkningsblokk. Behold eventuelle returnerte tenkningsblokker uendret i samtale
historikken.
Se
[Thinking](https://platform.claude.com/docs/en/build-with-claude/extended-thinking).
Eksemplene utelater `temperature`, `top_p` og `top_k`; se dokumentasjonen for den valgte
modellens parametere før du legger til samplingkontroller.
## Mellomlagre Prompt-er
Plasser et cache-breakpoint på referansemateriale som du gjenbruker mellom forespørsler.
Lagre referansematerialet ditt i en UTF-8-`reference.txt`-fil før du kjører dette
eksempelet. Bruk et prefiks som oppfyller den valgte modellens
[minste mellomlagringsbare lengde](https://platform.claude.com/docs/en/build-with-claude/prompt-caching#cache-limitations):
```python theme={null}
import os
from pathlib import Path
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
reference = Path("reference.txt").read_text(encoding="utf-8")
with client.messages.stream(
model="claude-sonnet-5",
max_tokens=1024,
messages=[
{
"role": "user",
"content": "How long can customers return damaged books, and what must they provide?",
},
],
system=[
{
"type": "text",
"text": reference,
"cache_control": {
"type": "ephemeral",
},
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
print(message.usage.model_dump_json())
```
Kjør forespørselen på nytt med samme modell, referansetekst og tenkningsinnstillinger.
Undersøk brukstellerne for å avgjøre om forespørselen gjenbrukte et prefiks:
* `cache_creation_input_tokens` teller tokens som skrives til cachen.
* `cache_read_input_tokens` teller tokens som leses fra cachen.
* `input_tokens` teller inndata som behandles utenfor disse cache-tellerne.
Ved gjentatte forespørsler rapporterer `cache_read_input_tokens` hvor mange input tokens
som ble lest fra cachen. OpenAPI- **Prompt Cache**
-eksempelet inneholder en komplett fiktiv referanse som du kan lagre som
`reference.txt` for å prøve eksempelet.
## Strøm svar
Angi `stream: true` for Server-Sent Events. SDK-en eksponerer tekstfragmenter etter hvert som de
ankommer:
```python theme={null}
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=1024,
messages=[
{
"role": "user",
"content": "Write a two-line poem about a comet.",
},
],
) as stream:
for text in stream.text_stream:
print(text, end="", flush=True)
```
En meldingsstrøm inneholder `message_start`, innholdsblokkhendelser, `message_delta`,
og `message_stop`. Innholdsblokker kan inneholde tekst, tenkning eller verktøyaktivitet.
For en tekstblokk inneholder `content_block_delta` en `text_delta`. Les endelige bruksdata
og stoppårsaken fra `message_delta`.
## Kontroller innsats
Angi `output_config.effort` for å styre mengden resonnering. Dette eksemplet bruker
`low` for en kort forklaring og venter på den fullførte strømmede meldingen:
```python theme={null}
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=1000,
messages=[
{
"role": "user",
"content": "Summarize the benefit of automated tests in one sentence.",
},
],
output_config={
"effort": "low",
},
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
```
Bruk den offisielle
[referansen for innsats](https://platform.claude.com/docs/en/build-with-claude/effort)
for å velge et innsatsnivå. Angi `max_tokens` separat for å begrense utdataenes lengde.
## Bruk serververktøy
Serververktøy kjøres under API-forespørselen og returnerer resultatblokker sammen med
Claudes svar.
Hent en artikkel og be Claude bruke det hentede dokumentet i svaret sitt.
Dette eksemplet streamer svaret og undersøker både den endelige teksten og
`web_fetch_tool_result` blokken:
```python theme={null}
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-sonnet-5",
max_tokens=2048,
messages=[
{
"role": "user",
"content": "Fetch https://arxiv.org/abs/1512.03385 and state the paper title and the proposed learning framework in two sentences.",
},
],
tools=[
{
"type": "web_fetch_20250910",
"name": "web_fetch",
"max_uses": 3,
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
for block in message.content:
if block.type == "web_fetch_tool_result":
print(block.model_dump_json())
```
Svaret kombinerer en `server_tool_use`-blokk med en `web_fetch_tool_result`
som inneholder det hentede dokumentet eller en verktøyfeil.
Søk etter en offisiell lanseringsdato og be Claude oppgi kilden sin.
Dette eksemplet streamer svaret, samler inn den endelige meldingen og skriver ut de
returnerte søkekildene:
```python theme={null}
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-sonnet-5",
max_tokens=4096,
messages=[
{
"role": "user",
"content": "Search the web for the official Python 3.14 release date. Answer with the date and cite a source.",
},
],
tools=[
{
"type": "web_search_20250305",
"name": "web_search",
"max_uses": 3,
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
for block in message.content:
if block.type == "web_search_tool_result":
if isinstance(block.content, list):
for result in block.content:
if result.type == "web_search_result":
print(result.title, result.url)
```
Svaret kombinerer en `server_tool_use`-blokk med en `web_search_tool_result`
som inneholder kilder. Les kildetitlene og URL-ene fra denne resultatblokken.
## Returner resultater fra klientverktøy
For et klientverktøy returnerer Claude en `tool_use`-blokk. Kjør applikasjonsfunksjonen
og send resultatet i en `tool_result`-blokk med den samsvarende `tool_use_id`.
Behold alt assistentinnholdet mellom de to forespørslene.
Dette eksemplet oppgir et fiktivt bestillingsresultat og ber Claude bruke dette resultatet:
```python theme={null}
import json
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
tools = [{
"name": "get_order_status",
"description": "Look up the status of a demonstration order.",
"input_schema": {
"type": "object",
"properties": {
"order_id": {
"type": "string",
"description": "The order reference.",
},
},
"required": ["order_id"],
},
}]
messages = [{
"role": "user",
"content": "Check the status of demo-order-42 using the order tool.",
}]
with client.messages.stream(
model="claude-opus-5",
max_tokens=1000,
messages=messages,
tools=tools,
) as stream:
message = stream.get_final_message()
tool_call = next(b for b in message.content if b.type == "tool_use")
messages.append({
"role": "assistant",
"content": [block.model_dump() for block in message.content],
})
messages.append({
"role": "user",
"content": [{
"type": "tool_result",
"tool_use_id": tool_call.id,
"content": json.dumps({
"status": "ready_for_pickup",
"pickup_code": "COMET42",
}),
}],
})
with client.messages.stream(
model="claude-opus-5",
max_tokens=1000,
messages=messages,
tools=tools,
) as stream:
follow_up = stream.get_final_message()
for block in follow_up.content:
if block.type == "text":
print(block.text)
```
## Svareksempel
En forespørsel uten strømming returnerer et meldingsobjekt. Følgende eksempel viser dets
tekst- og bruksfelt med en illustrativ meldingsidentifikator:
```json theme={null}
{
"id": "msg_example",
"type": "message",
"role": "assistant",
"model": "claude-opus-5",
"content": [
{
"type": "text",
"text": "Here are three search queries covering different angles of the topic:\n\n1. **\"grid-scale battery storage cost trends 2024 2025\"**\n — Targets economics and recent price data for lithium-ion and other utility-scale systems.\n\n2. **\"long-duration energy storage technologies comparison flow battery vs compressed air vs thermal\"**\n — Surfaces technical comparisons of options for storage beyond the 4–8 hour range, where lithium-ion becomes less cost-effective.\n\n3. **\"green hydrogen seasonal storage feasibility round-trip efficiency\"**\n — Digs into the viability of hydrogen for multi-week or seasonal balancing, including its efficiency penalties.\n\n**A few tips for refining these:**\n- Add `site:.gov` or `site:.edu` to prioritize research and agency reports (e.g., NREL, IEA, DOE).\n- Append a region like `Europe`, `India`, or `California` if you need location-specific policy or deployment data.\n- Try `filetype:pdf` to pull up technical white papers and government studies directly.\n\nWant me to tailor these toward a specific use case — academic research, investment analysis, or policy work?"
}
],
"stop_reason": "end_turn",
"stop_sequence": null,
"usage": {
"input_tokens": 23,
"cache_creation_input_tokens": 0,
"cache_read_input_tokens": 0,
"cache_creation": {
"ephemeral_5m_input_tokens": 0,
"ephemeral_1h_input_tokens": 0
},
"output_tokens": 437,
"output_tokens_details": {
"thinking_tokens": 50
}
}
}
```
Stoppårsaken beskriver neste handling. `end_turn` fullfører svaret,
`max_tokens` betyr at utdataene nådde grensen, og `tool_use` ber om et klientverktøy
resultat. For en serververktøyrunde som returnerer `pause_turn`, fortsett med det
returnerte assistentinnholdet uendret.
## OpenAPI
````yaml api/openapi/text/post-anthropic-messages.openapi.json POST /v1/messages
openapi: 3.1.0
info:
title: Anthropic Messages API
version: 1.0.0
servers:
- url: https://api.cometapi.com
security:
- apiKeyAuth: []
paths:
/v1/messages:
post:
summary: Anthropic Messages
description: >-
Send Claude requests in the Anthropic Messages format. The examples
cover text and image input, conversation history, adaptive thinking,
client tools, prompt caching, streaming, and server tools.
operationId: anthropic_messages
parameters:
- name: anthropic-version
in: header
required: false
description: The Anthropic API version to use. Defaults to `2023-06-01`.
schema:
type: string
default: '2023-06-01'
example: '2023-06-01'
- name: anthropic-beta
in: header
required: false
description: >-
Comma-separated feature identifiers required by a specific beta API
feature. Omit this header for the examples on this page.
schema:
type: string
requestBody:
required: true
content:
application/json:
schema:
type: object
required:
- model
- messages
- max_tokens
properties:
model:
type: string
description: >-
The Claude model to use. See the [Models
page](/overview/models) for available Claude model IDs.
example: claude-opus-5
default: claude-opus-5
messages:
type: array
description: >-
Conversation history. Use user and assistant messages with
text strings or content-block arrays. Return complete
assistant content blocks when continuing a tool call.
items:
type: object
required:
- role
- content
properties:
role:
type: string
description: The role of the message author.
enum:
- user
- assistant
content:
description: >-
The message content. Either a plain string or an array
of content blocks for multimodal input.
oneOf:
- type: string
- type: array
items:
type: object
properties:
type:
type: string
description: The content block type.
enum:
- text
- image
- document
- thinking
- redacted_thinking
- tool_use
- tool_result
- server_tool_use
- web_fetch_tool_result
- web_search_tool_result
text:
type: string
description: Text content (for `text` type blocks).
source:
type: object
description: >-
Source data for `image` or `document`
blocks.
properties:
type:
type: string
enum:
- base64
- url
description: >-
Image source type: `url` for a public
image URL, or `base64` for inline data.
media_type:
type: string
description: >-
Image MIME type for base64 sources, such
as `image/jpeg` or `image/png`.
data:
type: string
description: >-
Base64-encoded image bytes when `type`
is `base64`.
url:
type: string
description: >-
Public HTTPS URL for the image or
document.
cache_control:
type: object
description: >-
Cache control for prompt caching. Set
`{"type": "ephemeral"}` to cache this
content block.
properties:
type:
type: string
enum:
- ephemeral
description: Cache control type. Use `ephemeral`.
ttl:
type: string
enum:
- 5m
- 1h
description: >-
Cache lifetime. Use 5m for five minutes
or 1h for one hour. The default is 5m.
id:
type: string
description: >-
The ID from a returned tool_use block.
Preserve this value in assistant history.
name:
type: string
description: The function name in a tool_use block.
input:
type: object
description: Function arguments in a tool_use block.
tool_use_id:
type: string
description: >-
The ID of the tool_use block answered by
this tool_result.
content:
description: >-
Tool result content: a string, an array of
content blocks, or a server-tool result or
error object.
oneOf:
- type: string
- type: array
items:
type: object
- type: object
is_error:
type: boolean
description: >-
Set true when a client tool_result reports a
failed tool execution.
thinking:
type: string
description: >-
Thinking content returned in an assistant
block. Preserve it unchanged in conversation
history.
signature:
type: string
description: >-
Thinking signature returned by the API.
Preserve it unchanged with its thinking
block.
max_tokens:
type: integer
description: >-
The maximum number of tokens to generate. The model may stop
before reaching this limit. When using `thinking`, the
thinking tokens count towards this limit.
minimum: 1
example: 1024
system:
description: >-
System prompt providing context and instructions to Claude.
Can be a plain string or an array of content blocks (useful
for prompt caching).
oneOf:
- type: string
- type: array
items:
type: object
properties:
type:
type: string
enum:
- text
description: Content block type, `text`.
text:
type: string
description: System prompt text for this block.
cache_control:
type: object
properties:
type:
type: string
enum:
- ephemeral
description: Cache control type. Use `ephemeral`.
ttl:
type: string
enum:
- 5m
- 1h
description: >-
Cache lifetime. Use 5m for five minutes or 1h
for one hour. The default is 5m.
description: >-
Marks this system block as a prompt-cache
breakpoint.
temperature:
type: number
description: Sampling temperature. The examples omit sampling overrides.
minimum: 0
maximum: 1
top_p:
type: number
description: >-
Nucleus sampling threshold. The examples omit sampling
overrides.
minimum: 0
maximum: 1
top_k:
type: integer
description: >-
Limits sampling to the k most likely tokens. The examples
omit sampling overrides.
minimum: 0
stream:
type: boolean
description: >-
If `true`, stream the response incrementally using
Server-Sent Events (SSE). Events include `message_start`,
`content_block_start`, `content_block_delta`,
`content_block_stop`, `message_delta`, and `message_stop`.
default: false
stop_sequences:
type: array
description: >-
Custom strings that cause the model to stop generating when
encountered. The stop sequence is not included in the
response.
items:
type: string
thinking:
type: object
description: >-
Thinking configuration. The adaptive example uses type
adaptive and sets output_config.effort separately.
properties:
type:
type: string
description: >-
Thinking mode. Use adaptive for the example on this
page. Other values depend on the selected model.
enum:
- adaptive
- disabled
- enabled
budget_tokens:
type: integer
description: >-
Token budget for models that support manual thinking
with type enabled. Omit this field when using type
adaptive.
minimum: 1024
display:
type: string
description: >-
Controls visible thinking blocks on models that support
display selection. `summarized` returns a readable
thinking summary; `omitted` returns an empty thinking
field with continuity metadata where applicable.
enum:
- summarized
- omitted
tools:
type: array
description: >-
Client tools define a name and input_schema. Server tools
use a versioned type and name, such as the web_fetch and
web_search examples on this page.
items:
type: object
properties:
name:
type: string
description: >-
Tool name. The model repeats it in `tool_use` blocks
when it calls the tool.
description:
type: string
description: A description of what the tool does.
input_schema:
type: object
description: JSON Schema defining the tool's input parameters.
type:
type: string
description: >-
Versioned server tool type. The basic examples use
web_fetch_20250910 and web_search_20250305.
max_uses:
type: integer
description: Maximum uses of this server tool within one request.
tool_choice:
type: object
description: Controls how the model uses tools.
properties:
type:
type: string
description: >-
The tool choice mode: `auto` (model decides), `any`
(must use a tool), `tool` (must use a specific tool), or
`none` (no tools).
enum:
- auto
- any
- tool
- none
name:
type: string
description: >-
The specific tool name to use. Required when `type` is
`tool`.
disable_parallel_tool_use:
type: boolean
description: >-
If `true`, prevent the model from calling multiple tools
in parallel.
metadata:
type: object
description: Request metadata for tracking and analytics.
properties:
user_id:
type: string
description: >-
An external identifier for the user making the request.
Used for abuse detection.
output_config:
type: object
description: Configuration for reasoning effort and structured output.
properties:
effort:
type: string
description: >-
Reasoning effort. The examples use low for a brief
answer and xhigh with adaptive thinking.
enum:
- low
- medium
- high
- xhigh
- max
example: medium
format:
type: object
description: >-
Structured output configuration for supported models.
See the [structured outputs
documentation](https://platform.claude.com/docs/en/build-with-claude/structured-outputs)
for schema requirements.
required:
- type
- schema
properties:
type:
type: string
enum:
- json_schema
description: >-
Set json_schema to request output matching the
supplied schema.
schema:
type: object
description: JSON Schema that describes the requested output.
service_tier:
type: string
description: >-
The service tier to use. `auto` tries priority capacity
first, `standard_only` uses only standard capacity.
enum:
- auto
- standard_only
examples:
Default:
summary: Basic message
value:
model: claude-opus-5
max_tokens: 1000
messages:
- role: user
content: >-
Suggest three search queries about renewable energy
storage.
Image Input:
summary: Image Input
value:
model: claude-opus-5
max_tokens: 1000
messages:
- role: user
content:
- type: image
source:
type: url
url: https://picsum.photos/seed/comet/800/600.jpg
- type: text
text: Describe the visible scene in one sentence.
stream: true
Prompt Cache:
summary: With prompt caching
value:
model: claude-sonnet-5
max_tokens: 1024
messages:
- role: user
content: >-
How long can customers return damaged books, and what
must they provide?
system:
- type: text
text: >-
Comet Books sample reference. The following policies
describe a fictional bookstore.
Section 1: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 2: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 3: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 4: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 5: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 6: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 7: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 8: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 9: Damaged books may be returned within 30 days.
A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 10: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 11: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 12: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 13: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 14: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 15: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 16: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 17: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 18: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 19: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 20: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 21: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 22: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 23: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 24: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 25: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 26: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 27: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
Section 28: Damaged books may be returned within 30
days. A receipt is required. Unopened stationery may be
exchanged within 14 days. Digital downloads are final.
Staff record the order reference and explain the return
steps clearly.
cache_control:
type: ephemeral
stream: true
description: >-
The code samples read this reference material from a local
UTF-8 reference.txt file. Keep the cached prefix unchanged
when repeating the request.
Streaming:
summary: Streaming response
value:
model: claude-opus-5
max_tokens: 1024
messages:
- role: user
content: Write a two-line poem about a comet.
stream: true
Web Fetch:
summary: With web fetch tool
value:
model: claude-sonnet-5
max_tokens: 2048
messages:
- role: user
content: >-
Fetch https://arxiv.org/abs/1512.03385 and state the
paper title and the proposed learning framework in two
sentences.
tools:
- type: web_fetch_20250910
name: web_fetch
max_uses: 3
stream: true
Adaptive Thinking:
summary: With adaptive thinking and effort
value:
model: claude-opus-5
max_tokens: 4096
messages:
- role: user
content: >-
A shop sells notebooks for $7 each. Buying three earns a
$4 discount. A customer pays $20 for three notebooks.
Calculate the total and the change, and explain your
steps.
thinking:
type: adaptive
output_config:
effort: xhigh
stream: true
Effort control:
summary: Stream a concise response with low effort
value:
model: claude-opus-5
max_tokens: 1000
messages:
- role: user
content: >-
Summarize the benefit of automated tests in one
sentence.
output_config:
effort: low
stream: true
Web Search:
summary: Web Search
value:
model: claude-sonnet-5
max_tokens: 4096
messages:
- role: user
content: >-
Search the web for the official Python 3.14 release
date. Answer with the date and cite a source.
tools:
- type: web_search_20250305
name: web_search
max_uses: 3
stream: true
responses:
'200':
description: >-
Successful response. When `stream` is `true`, the response is a
stream of SSE events.
content:
application/json:
schema:
type: object
properties:
id:
type: string
description: Message identifier returned by the API.
type:
type: string
description: Always `message`.
enum:
- message
role:
type: string
description: Always `assistant`.
enum:
- assistant
content:
type: array
description: >-
The response content blocks. May include `text`,
`thinking`, `tool_use`, and other block types.
items:
type: object
properties:
type:
type: string
description: >-
Content block type, such as text, thinking,
tool_use, server_tool_use, web_fetch_tool_result, or
web_search_tool_result.
text:
type: string
description: The generated text (for `text` blocks).
thinking:
type: string
description: >-
Thinking text or summary for `thinking` blocks when
the selected model returns visible thinking content.
signature:
type: string
description: Cryptographic signature for the thinking block.
id:
type: string
description: Tool use ID (for `tool_use` blocks).
name:
type: string
description: Tool name (for `tool_use` blocks).
input:
type: object
description: Tool input arguments (for `tool_use` blocks).
tool_use_id:
type: string
description: The server tool call answered by this result block.
content:
description: >-
Server tool results or an error object. Inspect the
block type before reading its fields.
oneOf:
- type: object
- type: array
items:
type: object
citations:
type: array
description: >-
Source citations attached to a text block when
returned.
items:
type: object
model:
type: string
description: Model ID reported by the response.
stop_reason:
type: string
description: >-
Why the model stopped generating. `refusal` can be
returned as a successful HTTP response when the model
declines a request.
enum:
- end_turn
- max_tokens
- stop_sequence
- tool_use
- pause_turn
- refusal
stop_sequence:
type:
- string
- 'null'
description: >-
The stop sequence that caused the model to stop, if
applicable.
usage:
type: object
description: Token usage statistics.
properties:
input_tokens:
type: integer
description: >-
Uncached input tokens processed for the request. Cache
creation and cache reads have separate counters.
output_tokens:
type: integer
description: >-
Total output tokens, including thinking when the model
uses thinking.
cache_creation_input_tokens:
type: integer
description: >-
Number of input tokens used to create the prompt
cache.
cache_read_input_tokens:
type: integer
description: Number of input tokens read from the prompt cache.
cache_creation:
type: object
description: Detailed cache creation token breakdown by TTL tier.
properties:
ephemeral_5m_input_tokens:
type: integer
description: Tokens written to 5-minute ephemeral cache.
ephemeral_1h_input_tokens:
type: integer
description: Tokens written to 1-hour ephemeral cache.
output_tokens_details:
type: object
description: >-
Detailed output token breakdown when returned by the
selected model.
properties:
thinking_tokens:
type: integer
description: >-
Number of output tokens used for model thinking
when reported.
server_tool_use:
type: object
description: Server tool request counts when server tools are used.
properties:
web_fetch_requests:
type: integer
description: Number of web fetch requests.
web_search_requests:
type: integer
description: Number of web search requests.
example:
id: msg_example
type: message
role: assistant
model: claude-opus-5
content:
- type: text
text: >-
Here are three search queries covering different angles
of the topic:
1. **"grid-scale battery storage cost trends 2024
2025"**
— Targets economics and recent price data for lithium-ion and other utility-scale systems.
2. **"long-duration energy storage technologies
comparison flow battery vs compressed air vs thermal"**
— Surfaces technical comparisons of options for storage beyond the 4–8 hour range, where lithium-ion becomes less cost-effective.
3. **"green hydrogen seasonal storage feasibility
round-trip efficiency"**
— Digs into the viability of hydrogen for multi-week or seasonal balancing, including its efficiency penalties.
**A few tips for refining these:**
- Add `site:.gov` or `site:.edu` to prioritize research
and agency reports (e.g., NREL, IEA, DOE).
- Append a region like `Europe`, `India`, or
`California` if you need location-specific policy or
deployment data.
- Try `filetype:pdf` to pull up technical white papers
and government studies directly.
Want me to tailor these toward a specific use case —
academic research, investment analysis, or policy work?
stop_reason: end_turn
stop_sequence: null
usage:
input_tokens: 23
cache_creation_input_tokens: 0
cache_read_input_tokens: 0
cache_creation:
ephemeral_5m_input_tokens: 0
ephemeral_1h_input_tokens: 0
output_tokens: 437
output_tokens_details:
thinking_tokens: 50
example:
id: msg_example
type: message
role: assistant
model: claude-opus-5
content:
- type: text
text: >-
Here are three search queries covering different angles of
the topic:
1. **"grid-scale battery storage cost trends 2024 2025"**
— Targets economics and recent price data for lithium-ion and other utility-scale systems.
2. **"long-duration energy storage technologies comparison
flow battery vs compressed air vs thermal"**
— Surfaces technical comparisons of options for storage beyond the 4–8 hour range, where lithium-ion becomes less cost-effective.
3. **"green hydrogen seasonal storage feasibility
round-trip efficiency"**
— Digs into the viability of hydrogen for multi-week or seasonal balancing, including its efficiency penalties.
**A few tips for refining these:**
- Add `site:.gov` or `site:.edu` to prioritize research
and agency reports (e.g., NREL, IEA, DOE).
- Append a region like `Europe`, `India`, or `California`
if you need location-specific policy or deployment data.
- Try `filetype:pdf` to pull up technical white papers and
government studies directly.
Want me to tailor these toward a specific use case —
academic research, investment analysis, or policy work?
stop_reason: end_turn
stop_sequence: null
usage:
input_tokens: 23
cache_creation_input_tokens: 0
cache_read_input_tokens: 0
cache_creation:
ephemeral_5m_input_tokens: 0
ephemeral_1h_input_tokens: 0
output_tokens: 437
output_tokens_details:
thinking_tokens: 50
text/event-stream:
schema:
type: string
description: >-
Anthropic SSE events. A completed message ends with
message_stop; text_delta events contain text fragments.
x-codeSamples:
- lang: Python
label: Basic
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
message = client.messages.create(
model="claude-opus-5",
max_tokens=1000,
messages=[
{
"role": "user",
"content": "Suggest three search queries about renewable energy storage.",
},
],
)
for block in message.content:
if block.type == "text":
print(block.text)
- lang: Python
label: Prompt Cache
source: |
import os
from pathlib import Path
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
reference = Path("reference.txt").read_text(encoding="utf-8")
with client.messages.stream(
model="claude-sonnet-5",
max_tokens=1024,
messages=[
{
"role": "user",
"content": "How long can customers return damaged books, and what must they provide?",
},
],
system=[
{
"type": "text",
"text": reference,
"cache_control": {
"type": "ephemeral",
},
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
print(message.usage.model_dump_json())
- lang: Python
label: Adaptive thinking
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=4096,
messages=[
{
"role": "user",
"content": "A shop sells notebooks for $7 each. Buying three earns a $4 discount. A customer pays $20 for three notebooks. Calculate the total and the change, and explain your steps.",
},
],
thinking={
"type": "adaptive",
},
output_config={
"effort": "xhigh",
},
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
- lang: Python
label: Effort control
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=1000,
messages=[
{
"role": "user",
"content": "Summarize the benefit of automated tests in one sentence.",
},
],
output_config={
"effort": "low",
},
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
- lang: Python
label: Streaming
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=1024,
messages=[
{
"role": "user",
"content": "Write a two-line poem about a comet.",
},
],
) as stream:
for text in stream.text_stream:
print(text, end="", flush=True)
- lang: Python
label: Web Fetch
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-sonnet-5",
max_tokens=2048,
messages=[
{
"role": "user",
"content": "Fetch https://arxiv.org/abs/1512.03385 and state the paper title and the proposed learning framework in two sentences.",
},
],
tools=[
{
"type": "web_fetch_20250910",
"name": "web_fetch",
"max_uses": 3,
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
for block in message.content:
if block.type == "web_fetch_tool_result":
print(block.model_dump_json())
- lang: Python
label: Image Input
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-opus-5",
max_tokens=1000,
messages=[
{
"role": "user",
"content": [
{
"type": "image",
"source": {
"type": "url",
"url": "https://picsum.photos/seed/comet/800/600.jpg",
},
},
{
"type": "text",
"text": "Describe the visible scene in one sentence.",
},
],
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
- lang: JavaScript
label: Basic
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const message = await client.messages.create({
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "Suggest three search queries about renewable energy storage."
}
]
});
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
- lang: JavaScript
label: Prompt Cache
source: |
import Anthropic from "@anthropic-ai/sdk";
import fs from "node:fs";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const reference = fs.readFileSync("reference.txt", "utf8");
const stream = client.messages.stream({
"model": "claude-sonnet-5",
"max_tokens": 1024,
"messages": [
{
"role": "user",
"content": "How long can customers return damaged books, and what must they provide?"
}
],
"system": [
{
"type": "text",
"text": reference,
"cache_control": {
"type": "ephemeral"
}
}
]
});
const message = await stream.finalMessage();
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
console.log(message.usage);
- lang: JavaScript
label: Adaptive thinking
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const stream = client.messages.stream({
"model": "claude-opus-5",
"max_tokens": 4096,
"messages": [
{
"role": "user",
"content": "A shop sells notebooks for $7 each. Buying three earns a $4 discount. A customer pays $20 for three notebooks. Calculate the total and the change, and explain your steps."
}
],
"thinking": {
"type": "adaptive"
},
"output_config": {
"effort": "xhigh"
}
});
const message = await stream.finalMessage();
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
- lang: JavaScript
label: Effort control
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const stream = client.messages.stream({
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "Summarize the benefit of automated tests in one sentence."
}
],
"output_config": {
"effort": "low"
}
});
const message = await stream.finalMessage();
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
- lang: JavaScript
label: Streaming
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const stream = client.messages.stream({
"model": "claude-opus-5",
"max_tokens": 1024,
"messages": [
{
"role": "user",
"content": "Write a two-line poem about a comet."
}
]
});
stream.on("text", (text) => process.stdout.write(text));
await stream.finalMessage();
- lang: JavaScript
label: Web Fetch
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const stream = client.messages.stream({
"model": "claude-sonnet-5",
"max_tokens": 2048,
"messages": [
{
"role": "user",
"content": "Fetch https://arxiv.org/abs/1512.03385 and state the paper title and the proposed learning framework in two sentences."
}
],
"tools": [
{
"type": "web_fetch_20250910",
"name": "web_fetch",
"max_uses": 3
}
]
});
const message = await stream.finalMessage();
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
for (const block of message.content) {
if (block.type === "web_fetch_tool_result") console.log(block);
}
- lang: JavaScript
label: Image Input
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const stream = client.messages.stream({
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": [
{
"type": "image",
"source": {
"type": "url",
"url": "https://picsum.photos/seed/comet/800/600.jpg"
}
},
{
"type": "text",
"text": "Describe the visible scene in one sentence."
}
]
}
]
});
const message = await stream.finalMessage();
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
}
- lang: Shell
label: Basic
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "Suggest three search queries about renewable energy storage."
}
]
}'
- lang: Shell
label: Prompt Cache
source: |
request_body=$(jq -n --rawfile reference reference.txt '
{
"model": "claude-sonnet-5",
"max_tokens": 1024,
"messages": [
{
"role": "user",
"content": "How long can customers return damaged books, and what must they provide?"
}
],
"system": [
{
"type": "text",
"text": $reference,
"cache_control": {
"type": "ephemeral"
}
}
],
"stream": true
}
')
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
--data-binary "$request_body"
- lang: Shell
label: Adaptive thinking
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
-d '{
"model": "claude-opus-5",
"max_tokens": 4096,
"messages": [
{
"role": "user",
"content": "A shop sells notebooks for $7 each. Buying three earns a $4 discount. A customer pays $20 for three notebooks. Calculate the total and the change, and explain your steps."
}
],
"thinking": {
"type": "adaptive"
},
"output_config": {
"effort": "xhigh"
},
"stream": true
}'
- lang: Shell
label: Effort control
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
-d '{
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": "Summarize the benefit of automated tests in one sentence."
}
],
"output_config": {
"effort": "low"
},
"stream": true
}'
- lang: Shell
label: Streaming
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
-d '{
"model": "claude-opus-5",
"max_tokens": 1024,
"messages": [
{
"role": "user",
"content": "Write a two-line poem about a comet."
}
],
"stream": true
}'
- lang: Shell
label: Web Fetch
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
-d '{
"model": "claude-sonnet-5",
"max_tokens": 2048,
"messages": [
{
"role": "user",
"content": "Fetch https://arxiv.org/abs/1512.03385 and state the paper title and the proposed learning framework in two sentences."
}
],
"tools": [
{
"type": "web_fetch_20250910",
"name": "web_fetch",
"max_uses": 3
}
],
"stream": true
}'
- lang: Shell
label: Image Input
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
-d '{
"model": "claude-opus-5",
"max_tokens": 1000,
"messages": [
{
"role": "user",
"content": [
{
"type": "image",
"source": {
"type": "url",
"url": "https://picsum.photos/seed/comet/800/600.jpg"
}
},
{
"type": "text",
"text": "Describe the visible scene in one sentence."
}
]
}
],
"stream": true
}'
- lang: Python
label: Web Search
source: |
import os
import anthropic
client = anthropic.Anthropic(
base_url="https://api.cometapi.com",
api_key=os.environ["COMETAPI_KEY"],
)
with client.messages.stream(
model="claude-sonnet-5",
max_tokens=4096,
messages=[
{
"role": "user",
"content": "Search the web for the official Python 3.14 release date. Answer with the date and cite a source.",
},
],
tools=[
{
"type": "web_search_20250305",
"name": "web_search",
"max_uses": 3,
},
],
) as stream:
message = stream.get_final_message()
for block in message.content:
if block.type == "text":
print(block.text)
for block in message.content:
if block.type == "web_search_tool_result":
if isinstance(block.content, list):
for result in block.content:
if result.type == "web_search_result":
print(result.title, result.url)
- lang: JavaScript
label: Web Search
source: |
import Anthropic from "@anthropic-ai/sdk";
const client = new Anthropic({
apiKey: process.env.COMETAPI_KEY,
baseURL: "https://api.cometapi.com",
});
const stream = client.messages.stream({
"model": "claude-sonnet-5",
"max_tokens": 4096,
"messages": [
{
"role": "user",
"content": "Search the web for the official Python 3.14 release date. Answer with the date and cite a source."
}
],
"tools": [
{
"type": "web_search_20250305",
"name": "web_search",
"max_uses": 3
}
]
});
const message = await stream.finalMessage();
for (const block of message.content) {
if (block.type === "text") console.log(block.text);
if (block.type === "web_search_tool_result" && Array.isArray(block.content)) {
for (const result of block.content) {
if (result.type === "web_search_result") console.log(result.title, result.url);
}
}
}
- lang: Shell
label: Web Search
source: |
curl "https://api.cometapi.com/v1/messages" \
-H "Content-Type: application/json" \
-H "x-api-key: $COMETAPI_KEY" \
-H "anthropic-version: 2023-06-01" \
--no-buffer \
-d '{
"model": "claude-sonnet-5",
"max_tokens": 4096,
"messages": [
{
"role": "user",
"content": "Search the web for the official Python 3.14 release date. Answer with the date and cite a source."
}
],
"tools": [
{
"type": "web_search_20250305",
"name": "web_search",
"max_uses": 3
}
],
"stream": true
}'
components:
securitySchemes:
apiKeyAuth:
type: apiKey
in: header
name: x-api-key
description: >-
Your CometAPI key passed via the `x-api-key` header. `Authorization:
Bearer $COMETAPI_KEY` is also supported.
````