> ## 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.
# Tạo phản hồi model
> Sử dụng CometAPI POST /v1/responses để tạo các phản hồi model Multimodal có trạng thái với các công cụ tích hợp sẵn và Function Calling.
Responses API mở rộng [Chat Completions](/vi/api/text/chat) với các cuộc hội thoại có trạng thái, công cụ tích hợp sẵn, dữ liệu đầu vào tệp Multimodal và khả năng kiểm soát reasoning. Đây là endpoint được đề xuất cho các model reasoning OpenAI o-series, dòng GPT-5 và các model Codex.
Các nhà cung cấp model khác nhau hỗ trợ các tham số yêu cầu khác nhau và trả về các trường phản hồi khác nhau. Không phải mọi tham số được liệt kê trong playground ở trên đều hoạt động với mọi model trên CometAPI.
***
## Sử dụng các cuộc hội thoại có trạng thái
Liên kết các phản hồi với nhau bằng `previous_response_id` thay vì tự quản lý lịch sử tin nhắn:
```python theme={null}
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
# First turn
response = client.responses.create(
model="gpt-5.6-sol",
input="What is quantum computing?",
)
# Second turn — previous context is included automatically
follow_up = client.responses.create(
model="gpt-5.6-sol",
input="Can you explain that more simply?",
previous_response_id=response.id,
)
print(follow_up.output_text)
```
***
## Sử dụng các công cụ tích hợp sẵn
Responses API bao gồm các công cụ do nền tảng cung cấp, không cần cấu hình:
| Công cụ | Mục đích |
| -------------------- | ----------------------------------------------- |
| `web_search_preview` | Tìm kiếm thông tin theo thời gian thực trên web |
| `file_search` | Tìm kiếm trong các tệp đã tải lên |
| `code_interpreter` | Thực thi mã Python trong môi trường sandbox |
Để bật một công cụ tích hợp sẵn, hãy thêm công cụ đó vào mảng `tools`:
```python theme={null}
response = client.responses.create(
model="gpt-5.6-sol",
input="Find the current price of Bitcoin",
tools=[{"type": "web_search_preview"}],
)
print(response.output_text)
```
***
## Gọi các hàm tùy chỉnh
Xác định các hàm mà model có thể gọi bằng các đối số có cấu trúc:
```python theme={null}
response = client.responses.create(
model="gpt-5.6-sol",
input="What's the weather in Tokyo?",
tools=[{
"type": "function",
"name": "get_weather",
"description": "Get current weather for a location",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string"}
},
"required": ["location"]
}
}],
)
```
Khi model gọi một hàm, mảng `output` của phản hồi chứa một mục `function_call` với tên hàm và các đối số đã được phân tích cú pháp. Thực thi hàm rồi gửi kết quả trở lại trong một yêu cầu tiếp theo.
***
## Yêu cầu đầu ra có cấu trúc
Để buộc đầu ra JSON khớp với một schema cụ thể, hãy sử dụng tham số `text.format`:
```python theme={null}
response = client.responses.create(
model="gpt-5.6-sol",
input="List 3 programming languages with their main use cases",
text={
"format": {
"type": "json_schema",
"name": "languages",
"strict": True,
"schema": {
"type": "object",
"properties": {
"languages": {
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {"type": "string"},
"use_case": {"type": "string"}
},
"required": ["name", "use_case"],
"additionalProperties": False
}
}
},
"required": ["languages"],
"additionalProperties": False
}
}
},
)
```
***
## Cấu hình reasoning
Đối với các model o-series và GPT-5, hãy kiểm soát độ sâu reasoning bằng `reasoning.effort`:
```python theme={null}
response = client.responses.create(
model="o3",
input="Solve this step by step: if f(x) = x^3 - 6x^2 + 11x - 6, find all roots.",
reasoning={"effort": "high"}, # "low", "medium", or "high"
)
print(response.output_text)
```
Mức reasoning effort cao hơn tạo ra câu trả lời kỹ lưỡng hơn nhưng dùng nhiều Tokens hơn. Sử dụng `"low"` cho các truy vấn đơn giản và `"high"` cho các vấn đề phức tạp gồm nhiều bước.
***
## Streaming phản hồi
Để nhận đầu ra tăng dần, đặt `stream` thành `true`. API gửi các sự kiện do máy chủ gửi SSE theo thứ tự này:
1. `response.created` — Đối tượng phản hồi đã được khởi tạo
2. `response.in_progress` — Quá trình tạo đã bắt đầu
3. `response.output_item.added` — Mục đầu ra mới, gồm tin nhắn hoặc lệnh gọi công cụ)
4. `response.content_part.added` — Phần nội dung đã bắt đầu
5. `response.output_text.delta` — Đoạn văn bản chứa trường `delta`)
6. `response.output_text.done` — Hoàn tất tạo văn bản cho phần nội dung này
7. `response.content_part.done` — Phần nội dung đã hoàn tất
8. `response.output_item.done` — Mục đầu ra đã hoàn tất
9. `response.completed` — Phản hồi đầy đủ với dữ liệu `usage`
Streaming một phản hồi bằng Python SDK:
```python theme={null}
stream = client.responses.create(
model="gpt-5.6-sol",
input="Write a haiku about coding",
stream=True,
)
for event in stream:
if event.type == "response.output_text.delta":
print(event.delta, end="")
```
***
Để xem hướng dẫn chuyên sâu về từng khả năng, hãy tham khảo tài liệu OpenAI:
[Văn bản](https://developers.openai.com/docs/guides/text) · [Hình ảnh](https://developers.openai.com/docs/guides/images) · [Tệp PDF](https://developers.openai.com/docs/guides/pdf-files) · [Đầu ra có cấu trúc](https://developers.openai.com/docs/guides/structured-outputs) · [Function Calling](https://developers.openai.com/docs/guides/function-calling) · [Trạng thái hội thoại](https://developers.openai.com/docs/guides/conversation-state) · [Công cụ tích hợp sẵn](https://developers.openai.com/docs/guides/tools) · [Reasoning](https://developers.openai.com/docs/guides/reasoning)
## OpenAPI
````yaml api/openapi/text/post-responses.openapi.json POST /v1/responses
openapi: 3.1.0
info:
title: Responses API
version: 1.0.0
servers:
- url: https://api.cometapi.com
security:
- bearerAuth: []
paths:
/v1/responses:
post:
summary: Create Response
operationId: createResponse
requestBody:
required: true
content:
application/json:
schema:
type: object
required:
- model
- input
properties:
model:
type: string
description: >-
Model ID to use for this request. See the [Models
page](/overview/models) for current options.
example: gpt-5.6-sol
input:
oneOf:
- type: string
description: A plain text string as input.
- type: array
description: >-
An array of input items with roles and multimodal
content.
items:
type: object
description: >-
Text, image, or file inputs to the model, used to generate a
response. Can be a simple string for text-only input, or an
array of input items for multimodal content (images, files)
and multi-turn conversations.
instructions:
type: string
description: >-
A system (or developer) message inserted into the model's
context. When used with `previous_response_id`, instructions
from the previous response are not carried over — this makes
it easy to swap system messages between turns.
background:
type: boolean
description: >-
Whether to run the model response in the background.
Background responses do not return output directly — you
retrieve the result later via the response ID.
default: false
context_management:
type: array
description: >-
Context management configuration for this request. Controls
how the model manages context when the conversation exceeds
the context window.
items:
type: object
properties:
type:
type: string
description: The type of context management.
compact_threshold:
type: number
description: >-
The threshold at which context compaction is
triggered.
conversation:
type:
- string
- object
- 'null'
description: >-
The conversation this response belongs to. Items from the
conversation are prepended to `input` for context. Input and
output items are automatically added to the conversation
after the response completes. Cannot be used with
`previous_response_id`.
default: null
include:
type: array
description: >-
Additional output data to include in the response. Use this
to request extra information that is not included by
default.
items:
type: string
enum:
- web_search_call.action.sources
- code_interpreter_call.outputs
- computer_call_output.output.image_url
- file_search_call.results
- message.input_image.image_url
- message.output_text.logprobs
- reasoning.encrypted_content
max_output_tokens:
type: integer
description: >-
An upper bound for the number of tokens that can be
generated for a response, including visible output tokens
and reasoning tokens.
max_tool_calls:
type: integer
description: >-
The maximum number of total calls to built-in tools that can
be processed in a response. This limit applies across all
built-in tool calls, not per individual tool. Any further
tool call attempts by the model will be ignored.
metadata:
type: object
description: >-
Set of up to 16 key-value pairs that can be attached to the
response. Useful for storing additional information in a
structured format. Keys have a maximum length of 64
characters; values have a maximum length of 512 characters.
additionalProperties:
type: string
parallel_tool_calls:
type: boolean
description: Whether to allow the model to run tool calls in parallel.
default: true
previous_response_id:
type: string
description: >-
The unique ID of a previous response. Use this to create
multi-turn conversations without manually managing
conversation state. Cannot be used with `conversation`.
prompt:
type: object
description: Reference to a prompt template and its variables.
properties:
id:
type: string
description: The ID of the prompt template.
variables:
type: object
description: Key-value pairs for template variables.
additionalProperties:
type: string
version:
type: string
description: The version of the prompt template to use.
prompt_cache_key:
type: string
description: >-
A key used to cache responses for similar requests, helping
optimize cache hit rates. Replaces the deprecated `user`
field for caching purposes.
prompt_cache_retention:
type: string
description: >-
The retention policy for the prompt cache. Set to `24h` to
keep cached prefixes active for up to 24 hours.
enum:
- in-memory
- 24h
reasoning:
type: object
description: >-
Configuration options for reasoning models (o-series and
gpt-5). Controls the depth of reasoning before generating a
response.
properties:
effort:
type: string
description: >-
Constrains effort on reasoning. Reducing reasoning
effort can result in faster responses and fewer tokens
used on reasoning.
enum:
- none
- minimal
- low
- medium
- high
generate_summary:
type: string
description: Whether to generate a summary of the reasoning process.
enum:
- auto
- concise
- detailed
summary:
type: string
description: Deprecated. Use `generate_summary` instead.
enum:
- auto
- concise
- detailed
deprecated: true
safety_identifier:
type: string
description: >-
A stable identifier for your end-users, used to help detect
policy violations. Should be a hashed username or email — do
not send identifying information directly.
maxLength: 64
service_tier:
type: string
description: >-
Specifies the processing tier for the request. When set, the
response will include the actual `service_tier` used.
- `auto`: Uses the tier configured in project settings
(default behavior).
- `default`: Standard pricing and performance.
- `flex`: Flexible processing with potential cost savings.
- `priority`: Priority processing with faster response
times.
enum:
- auto
- default
- flex
- priority
store:
type: boolean
description: >-
Whether to store the generated response for later retrieval
via API.
default: true
stream:
type: boolean
description: >-
If set to `true`, the response data will be streamed to the
client as it is generated using server-sent events (SSE).
Events include `response.created`,
`response.output_text.delta`, `response.completed`, and
more.
default: false
stream_options:
type: object
description: >-
Options for streaming responses. Only set this when `stream`
is `true`.
properties:
include_obfuscation:
type: boolean
description: Whether to include obfuscation data in streaming events.
temperature:
type: number
description: >-
Sampling temperature between 0 and 2. Higher values (e.g.,
0.8) increase randomness; lower values (e.g., 0.2) make
output more focused and deterministic. We recommend
adjusting either this or `top_p`, but not both.
default: 1
minimum: 0
maximum: 2
text:
type: object
description: >-
Configuration for text output. Use this to request
structured JSON output via JSON mode or JSON Schema.
properties:
format:
type: object
description: The format of the text output.
properties:
type:
type: string
description: >-
The output format type. `text` returns plain text,
`json_object` returns valid JSON, `json_schema`
returns JSON conforming to a provided schema.
enum:
- text
- json_object
- json_schema
json_schema:
type: object
description: >-
The JSON Schema to use when `type` is `json_schema`.
Required when using structured outputs.
verbosity:
type: string
description: Controls the verbosity of the text output.
enum:
- low
- medium
- high
tool_choice:
type:
- string
- object
description: >-
Controls how the model selects which tool(s) to call.
- `auto` (default): The model decides whether and which
tools to call.
- `none`: The model will not call any tools.
- `required`: The model must call at least one tool.
- An object specifying a particular tool to use.
default: auto
tools:
type: array
description: >-
An array of tools the model may call while generating a
response. CometAPI supports three categories:
- **Built-in tools**: Platform-provided tools like
`web_search_preview` and `file_search`.
- **Function calls**: Custom functions you define, enabling
the model to call your own code with structured arguments.
- **MCP tools**: Integrations with third-party systems via
MCP servers.
items:
type: object
top_logprobs:
type: integer
description: >-
Number of most likely tokens to return at each position
(0–20), each with an associated log probability. Must
include `message.output_text.logprobs` in the `include`
parameter to receive logprobs.
minimum: 0
maximum: 20
top_p:
type: number
description: >-
Nucleus sampling parameter. The model considers tokens with
`top_p` cumulative probability mass. For example, 0.1 means
only the top 10% probability tokens are considered. We
recommend adjusting either this or `temperature`, but not
both.
default: 1
minimum: 0
maximum: 1
truncation:
type: string
description: >-
The truncation strategy for handling inputs that exceed the
model's context window.
- `auto`: The model truncates the input by dropping items
from the beginning of the conversation to fit.
- `disabled` (default): The request fails with a 400 error
if the input exceeds the context window.
enum:
- auto
- disabled
default: disabled
user:
type: string
description: >-
Deprecated. Use `safety_identifier` and `prompt_cache_key`
instead. A stable identifier for your end-user.
deprecated: true
examples:
Text Input:
summary: Text Input
value:
model: gpt-5.6-sol
input: Tell me a three sentence bedtime story about a unicorn.
Image Input:
summary: Image Input
value:
model: gpt-5.6-sol
input:
- role: user
content:
- type: input_text
text: What is in this image?
- type: input_image
image_url: >-
https://images.unsplash.com/photo-1500530855697-b586d89ba3ee?w=1200
File Input:
summary: File Input
value:
model: gpt-5.6-sol
input:
- role: user
content:
- type: input_text
text: What is in this file?
- type: input_file
file_url: >-
https://www.berkshirehathaway.com/letters/2024ltr.pdf
Function Calling:
summary: Function Calling
value:
model: gpt-5.6-sol
input: What is the weather like in Boston today?
tools:
- type: function
name: get_current_weather
description: Get the current weather in a given location
parameters:
type: object
properties:
location:
type: string
description: The city and state, e.g. San Francisco, CA
unit:
type: string
enum:
- celsius
- fahrenheit
description: Temperature unit.
required:
- location
- unit
Reasoning:
summary: Reasoning
value:
model: o4-mini
input: 'Solve step by step: What is 15% of 240?'
reasoning:
effort: high
responses:
'200':
description: The generated Response object.
content:
application/json:
schema:
type: object
properties:
id:
type: string
description: Unique identifier for the response.
example: resp_0a153ae8201f73bc0069a7e8044cc481
object:
type: string
description: The object type, always `response`.
enum:
- response
example: response
created_at:
type: integer
description: >-
Unix timestamp (in seconds) of when the response was
created.
example: 1772611588
status:
type: string
description: The status of the response.
enum:
- completed
- in_progress
- failed
- cancelled
- queued
example: completed
background:
type: boolean
description: Whether the response was run in the background.
example: false
completed_at:
type:
- integer
- 'null'
description: >-
Unix timestamp of when the response was completed, or
`null` if still in progress.
example: 1772611589
error:
type:
- object
- 'null'
description: >-
Error information if the response failed, or `null` on
success.
properties:
code:
type: string
description: The error code.
message:
type: string
description: A human-readable error message.
incomplete_details:
type:
- object
- 'null'
description: >-
Details about why the response is incomplete, if
applicable.
properties:
reason:
type: string
description: The reason the response is incomplete.
enum:
- max_output_tokens
- content_filter
instructions:
type:
- string
- 'null'
description: The system instructions used for this response.
max_output_tokens:
type:
- integer
- 'null'
description: The maximum output token limit that was applied.
model:
type: string
description: The model used for the response.
example: gpt-4.1-nano
output:
type: array
description: >-
An array of output items generated by the model. Each item
can be a message, function call, or other output type.
items:
type: object
properties:
id:
type: string
description: Unique identifier for the output item.
type:
type: string
description: The type of output item.
enum:
- message
- function_call
- web_search_call
- file_search_call
- code_interpreter_call
- computer_call
- reasoning
status:
type: string
description: The status of this output item.
enum:
- completed
- in_progress
role:
type: string
description: >-
The role of the message (present when `type` is
`message`).
enum:
- assistant
content:
type: array
description: >-
The content parts of the message (present when
`type` is `message`).
items:
type: object
properties:
type:
type: string
description: The content type.
enum:
- output_text
text:
type: string
description: The generated text content.
annotations:
type: array
description: >-
Annotations such as file citations or URL
citations.
items:
type: object
logprobs:
type: array
description: >-
Log probability information (when requested
via `include`).
items:
type: object
name:
type: string
description: >-
The name of the function being called (present when
`type` is `function_call`).
arguments:
type: string
description: >-
The JSON-encoded arguments for the function call
(present when `type` is `function_call`).
call_id:
type: string
description: >-
The unique call identifier (present when `type` is
`function_call`).
output_text:
type: string
description: >-
A convenience field containing the concatenated text
output from all output message items.
parallel_tool_calls:
type: boolean
description: Whether parallel tool calls were enabled.
previous_response_id:
type:
- string
- 'null'
description: >-
The ID of the previous response, if this is a multi-turn
conversation.
reasoning:
type: object
description: The reasoning configuration that was used.
properties:
effort:
type:
- string
- 'null'
description: The reasoning effort level.
summary:
type:
- string
- 'null'
description: The reasoning summary setting.
service_tier:
type: string
description: The service tier actually used to process the request.
example: default
store:
type: boolean
description: Whether the response was stored.
temperature:
type: number
description: The temperature value used.
example: 1
text:
type: object
description: The text configuration used.
properties:
format:
type: object
properties:
type:
type: string
description: >-
Format type: `text` (default), `json_object`, or
`json_schema`.
description: Output text format configuration.
verbosity:
type: string
description: The verbosity level used.
tool_choice:
type:
- string
- object
description: The tool choice setting used.
tools:
type: array
description: The tools that were available for this response.
items:
type: object
top_p:
type: number
description: The `top_p` value used.
example: 1
truncation:
type: string
description: The truncation strategy used.
usage:
type: object
description: Token usage statistics for this response.
properties:
input_tokens:
type: integer
description: Number of input tokens consumed.
example: 19
input_tokens_details:
type: object
description: Breakdown of input token usage.
properties:
cached_tokens:
type: integer
description: Number of input tokens that were cached.
example: 0
output_tokens:
type: integer
description: Number of output tokens generated.
example: 9
output_tokens_details:
type: object
description: Breakdown of output token usage.
properties:
reasoning_tokens:
type: integer
description: Number of tokens used for reasoning.
example: 0
total_tokens:
type: integer
description: Total number of tokens (input + output).
example: 28
user:
type:
- string
- 'null'
description: The user identifier, if provided.
metadata:
type: object
description: The metadata attached to this response.
additionalProperties:
type: string
content_filters:
type:
- array
- 'null'
description: Content filter results applied to the response, if any.
nullable: true
frequency_penalty:
type: number
description: The frequency penalty applied to the request.
default: 0
max_tool_calls:
type:
- integer
- 'null'
description: Maximum number of tool calls allowed, if set.
nullable: true
presence_penalty:
type: number
description: The presence penalty applied to the request.
default: 0
prompt_cache_key:
type:
- string
- 'null'
description: Cache key for prompt caching, if applicable.
nullable: true
prompt_cache_retention:
type:
- string
- 'null'
description: Prompt cache retention policy, if applicable.
nullable: true
safety_identifier:
type:
- string
- 'null'
description: Safety system identifier for the response, if applicable.
nullable: true
top_logprobs:
type: integer
description: >-
Number of top log probabilities returned per token
position.
default: 0
example:
id: resp_0a153ae8201f73bc0069a7e8044cc481
object: response
created_at: 1772611588
status: completed
background: false
completed_at: 1772611589
error: null
incomplete_details: null
instructions: null
max_output_tokens: null
model: gpt-4.1-nano
output:
- id: msg_0a153ae8201f73bc0069a7e8049a8881
type: message
status: completed
content:
- type: output_text
annotations: []
text: Four.
role: assistant
parallel_tool_calls: true
previous_response_id: null
prompt_cache_key: null
prompt_cache_retention: null
reasoning:
effort: null
summary: null
safety_identifier: null
service_tier: auto
store: true
temperature: 1
text:
format:
type: text
verbosity: medium
tool_choice: auto
tools: []
top_p: 1
truncation: disabled
usage:
input_tokens: 19
input_tokens_details:
cached_tokens: 0
output_tokens: 9
output_tokens_details:
reasoning_tokens: 0
total_tokens: 28
user: null
metadata: {}
x-codeSamples:
- lang: Python
label: Text Input
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
response = client.responses.create(
model="gpt-5.6-sol",
input="Tell me a three sentence bedtime story about a unicorn.",
)
print(response.output_text)
- lang: Python
label: Image Input
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
response = client.responses.create(
model="gpt-5.6-sol",
input=[
{
"role": "user",
"content": [
{"type": "input_text", "text": "What is in this image?"},
{
"type": "input_image",
"image_url": "https://images.unsplash.com/photo-1500530855697-b586d89ba3ee?w=1200",
},
],
}
],
)
print(response.output_text)
- lang: Python
label: File Input
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
response = client.responses.create(
model="gpt-5.6-sol",
input=[
{
"role": "user",
"content": [
{"type": "input_text", "text": "What is in this file?"},
{
"type": "input_file",
"file_url": "https://www.berkshirehathaway.com/letters/2024ltr.pdf",
},
],
}
],
)
print(response.output_text)
- lang: Python
label: Web Search
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
response = client.responses.create(
model="gpt-5.6-sol",
tools=[{"type": "web_search_preview"}],
input="What was a positive news story from today?",
)
print(response.output_text)
- lang: Python
label: Streaming
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
stream = client.responses.create(
model="gpt-5.6-sol",
input="Tell me a three sentence bedtime story about a unicorn.",
stream=True,
)
for event in stream:
print(event)
- lang: Python
label: Functions
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
response = client.responses.create(
model="gpt-5.6-sol",
input="What is the weather like in Boston today?",
tools=[
{
"type": "function",
"name": "get_current_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA",
},
"unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
},
"required": ["location", "unit"],
},
}
],
)
print(response.output)
- lang: Python
label: Reasoning
source: |
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.cometapi.com/v1",
api_key=os.environ["COMETAPI_KEY"],
)
response = client.responses.create(
model="o4-mini",
input="How much wood would a woodchuck chuck?",
reasoning={"effort": "high"},
)
print(response.output_text)
- lang: JavaScript
label: Text Input
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const response = await client.responses.create({
model: "gpt-5.6-sol",
input: "Tell me a three sentence bedtime story about a unicorn.",
});
console.log(response.output_text);
- lang: JavaScript
label: Image Input
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const response = await client.responses.create({
model: "gpt-5.6-sol",
input: [
{
role: "user",
content: [
{ type: "input_text", text: "What is in this image?" },
{
type: "input_image",
image_url: "https://images.unsplash.com/photo-1500530855697-b586d89ba3ee?w=1200",
},
],
},
],
});
console.log(response.output_text);
- lang: JavaScript
label: File Input
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const response = await client.responses.create({
model: "gpt-5.6-sol",
input: [
{
role: "user",
content: [
{ type: "input_text", text: "What is in this file?" },
{
type: "input_file",
file_url: "https://www.berkshirehathaway.com/letters/2024ltr.pdf",
},
],
},
],
});
console.log(response.output_text);
- lang: JavaScript
label: Web Search
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const response = await client.responses.create({
model: "gpt-5.6-sol",
tools: [{ type: "web_search_preview" }],
input: "What was a positive news story from today?",
});
console.log(response.output_text);
- lang: JavaScript
label: Streaming
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const stream = await client.responses.create({
model: "gpt-5.6-sol",
input: "Tell me a three sentence bedtime story about a unicorn.",
stream: true,
});
for await (const event of stream) {
if (event.type === "response.output_text.delta") {
process.stdout.write(event.delta);
}
}
- lang: JavaScript
label: Functions
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const response = await client.responses.create({
model: "gpt-5.6-sol",
input: "What is the weather like in Boston today?",
tools: [
{
type: "function",
name: "get_current_weather",
description: "Get the current weather in a given location",
parameters: {
type: "object",
properties: {
location: { type: "string", description: "The city and state" },
unit: { type: "string", enum: ["celsius", "fahrenheit"] },
},
required: ["location", "unit"],
},
},
],
});
console.log(response.output);
- lang: JavaScript
label: Reasoning
source: |
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.cometapi.com/v1",
apiKey: process.env.COMETAPI_KEY,
});
const response = await client.responses.create({
model: "o4-mini",
input: "How much wood would a woodchuck chuck?",
reasoning: { effort: "high" },
});
console.log(response.output_text);
- lang: Shell
label: Text Input
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "gpt-5.6-sol",
"input": "Tell me a three sentence bedtime story about a unicorn."
}'
- lang: Shell
label: Image Input
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "gpt-5.6-sol",
"input": [
{
"role": "user",
"content": [
{"type": "input_text", "text": "What is in this image?"},
{"type": "input_image", "image_url": "https://images.unsplash.com/photo-1500530855697-b586d89ba3ee?w=1200"}
]
}
]
}'
- lang: Shell
label: File Input
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "gpt-5.6-sol",
"input": [
{
"role": "user",
"content": [
{"type": "input_text", "text": "What is in this file?"},
{"type": "input_file", "file_url": "https://www.berkshirehathaway.com/letters/2024ltr.pdf"}
]
}
]
}'
- lang: Shell
label: Web Search
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "gpt-5.6-sol",
"tools": [{"type": "web_search_preview"}],
"input": "What was a positive news story from today?"
}'
- lang: Shell
label: Streaming
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "gpt-5.6-sol",
"input": "Tell me a three sentence bedtime story about a unicorn.",
"stream": true
}'
- lang: Shell
label: Functions
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "gpt-5.6-sol",
"input": "What is the weather like in Boston today?",
"tools": [
{
"type": "function",
"name": "get_current_weather",
"description": "Get the current weather in a given location",
"parameters": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city and state"},
"unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}
},
"required": ["location", "unit"]
}
}
]
}'
- lang: Shell
label: Reasoning
source: |
curl https://api.cometapi.com/v1/responses \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $COMETAPI_KEY" \
-d '{
"model": "o4-mini",
"input": "How much wood would a woodchuck chuck?",
"reasoning": {"effort": "high"}
}'
components:
securitySchemes:
bearerAuth:
type: http
scheme: bearer
description: Bearer token authentication. Use your CometAPI key.
````