Chat Models API Reference

This is the API reference for chat model endpoints. Additional non-chat API groups will be appended later.

Default Base URL: https://api.valueapi.ai/v1
Default Auth: Authorization: Bearer <API_KEY>
Note: Gemini native endpoint uses https://api.valueapi.ai/v1beta/models/....

Model Info (planned)Image (planned)Video (planned)Audio (planned)Music (planned)Realtime (planned)Tools (planned)

POST/v1/chat/completions

Chat API (General)

OpenAI-compatible general chat endpoint.

Works across OpenAI/Claude/Gemini style chat models.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Model ID.
messages	array<object>	Yes	Conversation history.
temperature	number	No	Sampling temperature.
top_p	number	No	Nucleus sampling control.
max_tokens	number	No	Maximum output tokens.
stream	boolean	No	SSE streaming switch.
response_format	object	No	Structured output options.
stop	string \| string[]	No	Stop sequences.
user	string	No	End-user identifier.

Body demo

{
  "model": "gpt-4o",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ],
  "temperature": 0.7,
  "max_tokens": 512,
  "stream": false
}

POST/v1/chat/completions

Chat API (Image Analysis)

Multimodal chat with text + image blocks.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Vision-capable model.
messages	array<object>	Yes	Message list.
messages[].content[]	array<object>	Yes	Mixed blocks.
messages[].content[].type	string	Yes	text \| image_url.
messages[].content[].text	string	No	Text prompt.
messages[].content[].image_url.url	string	No	Image URL or base64 data URL.
max_tokens	number	No	Maximum output tokens.

Body demo

{
  "model": "gpt-4o",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Describe this image"
        },
        {
          "type": "image_url",
          "image_url": {
            "url": "https://example.com/images/demo.jpg"
          }
        }
      ]
    }
  ],
  "max_tokens": 512
}

POST/v1/chat/completions

Chat API (Function Calling)

OpenAI-style tools/function calling.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Model ID.
messages	array<object>	Yes	Conversation messages.
tools	array<object>	Yes	Tool definitions list.
tools[].type	string	Yes	Must be function.
tools[].function.name	string	Yes	Function name.
tools[].function.parameters	object	Yes	JSON schema for arguments.
tools[].function.strict	boolean	No	Strict schema mode.

Body demo

{
  "model": "gpt-4o",
  "messages": [
    {
      "role": "user",
      "content": "What's weather in Paris?"
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get current weather by location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string"
            }
          },
          "required": ["location"]
        }
      }
    }
  ],
  "tool_choice": "auto"
}

POST/v1/chat/completions

Chat API (o1-o3 Models)

Special request shape for o1/o3 style models.

Source note: o1 stream is currently unsupported.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Example: o1-mini.
messages	array<object>	Yes	Conversation messages.
max_completion_tokens	number	No	Max generated tokens.
stream	boolean	No	Use false.

Body demo

{
  "model": "o1-mini",
  "messages": [
    {
      "role": "user",
      "content": "Explain this code"
    }
  ],
  "max_completion_tokens": 1024,
  "stream": false
}

POST/v1/chat/completions

gpt-4o-all File Analysis

Model alias for image/file/web-capable use cases.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Use gpt-4o-all.
messages	array<object>	Yes	Conversation messages.
messages[].content	string	Yes	Supports file_url + space + question.
max_tokens	number	No	Maximum output tokens.
temperature	number	No	Sampling temperature.

Body demo

{
  "model": "gpt-4o-all",
  "messages": [
    {
      "role": "user",
      "content": "https://example.com/files/api-doc.pdf Summarize this file"
    }
  ],
  "max_tokens": 1024,
  "temperature": 0.2
}

POST/v1/chat/completions

gpt-4-all File Analysis

GPT-4 based alias with file-analysis style input.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Use gpt-4-all.
messages	array<object>	Yes	Conversation messages.
messages[].content	string	Yes	Supports file_url + space + question.
max_tokens	number	No	Maximum output tokens.
temperature	number	No	Sampling temperature.

Body demo

{
  "model": "gpt-4-all",
  "messages": [
    {
      "role": "user",
      "content": "https://example.com/files/api-doc.pdf Summarize this file"
    }
  ],
  "max_tokens": 1024,
  "temperature": 0.2
}

POST/v1/completions

Legacy Text Completions

Classic prompt-completion endpoint.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Example: gpt-3.5-turbo-instruct.
prompt	string	Yes	Input prompt text.
max_tokens	number	No	Maximum output tokens.
temperature	number	No	Sampling temperature.
top_p	number	No	Nucleus sampling control.
frequency_penalty	number	No	Frequency penalty.
presence_penalty	number	No	Presence penalty.
logprobs	number	No	Top token probabilities count.

Body demo

{
  "model": "gpt-3.5-turbo-instruct",
  "prompt": "The weather is good",
  "max_tokens": 100,
  "temperature": 0.7,
  "top_p": 1
}

POST/v1/chat/completions

Claude (OpenAI Format)

Claude via OpenAI-compatible payload, including PDF/image-style blocks.

Source note: use native format for prompt caching behavior.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Claude model ID.
messages	array<object>	Yes	Conversation messages.
messages[].content[]	array<object>	No	Supports text/file/file_url/image_url blocks.
max_tokens	number	No	Maximum output tokens.
stream	boolean	No	Streaming switch.
thinking	object	No	Thinking config for supported Claude models.

Body demo

{
  "model": "claude-3-5-sonnet-20241022",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Summarize this PDF"
        },
        {
          "type": "file_url",
          "file_url": {
            "url": "https://example.com/files/api-doc.pdf"
          }
        }
      ]
    }
  ],
  "max_tokens": 1024,
  "stream": false
}

POST/v1/messages

Claude (Native Format)

Claude native endpoint with native content blocks and cache controls.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/messages

Body Parameters

Name	Type	Required	Description
model	string	Yes	Native-capable model ID.
messages	array<object>	Yes	Conversation messages.
messages[].content[]	array<object>	No	text/image/document blocks.
messages[].content[].source	object	No	base64 or url source object.
messages[].content[].cache_control	object	No	Optional cache config.
max_tokens	number	No	Maximum output tokens.
thinking	object	No	Thinking config on supported models.

Body demo

{
  "model": "claude-3-5-sonnet-20240620",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Summarize this PDF"
        },
        {
          "type": "document",
          "source": {
            "type": "url",
            "url": "https://example.com/files/api-doc.pdf"
          }
        }
      ]
    }
  ],
  "max_tokens": 1024
}

POST/v1/chat/completions

Gemini (OpenAI Format)

Gemini via OpenAI format with file analysis and search-style options.

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Gemini model ID.
messages	array<object>	Yes	Conversation messages.
messages[].content[]	array<object>	No	text/file/file_url/image_url blocks.
max_tokens	number	No	Maximum output tokens.
tools	array<object>	No	Function tools (e.g. googleSearch).
reasoning_effort	string	No	low \| medium \| high for compatible models.
extra_body.google.thinking_config	object	No	Gemini thinking controls.

Body demo

{
  "model": "gemini-2.5-pro",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Summarize this"
        },
        {
          "type": "file",
          "file": {
            "filename": "api-doc.pdf",
            "file_data": "data:application/pdf;base64,JVBER..."
          }
        }
      ]
    }
  ],
  "reasoning_effort": "medium",
  "max_tokens": 1024
}

POST/v1beta/models/{model}:{action}

Gemini (Native Format)

Gemini native endpoint using generateContent / streamGenerateContent.

Source note: prefer lowerCamelCase field names.

Headers

Content-Type: application/json
x-goog-api-key: <API_KEY> (or Bearer auth)

Full URL

https://api.valueapi.ai/v1beta/models/{model}:{action}

Body Parameters

Name	Type	Required	Description
contents	array<object>	Yes	Conversation content list.
contents[].role	string	Yes	Role, e.g. user.
contents[].parts	array<object>	Yes	Parts array.
contents[].parts[].text	string	Yes	Text content.

Body demo

{
  "contents": [
    {
      "role": "user",
      "parts": [
        {
          "text": "Hello"
        }
      ]
    }
  ],
  "generationConfig": {
    "temperature": 0.7
  }
}

POST/v1/chat/completions

GPTs

Call GPTs by model naming pattern gpt-4-gizmo-(gizmo_id).

Headers

Authorization: Bearer <API_KEY>
Content-Type: application/json

Full URL

https://api.valueapi.ai/v1/chat/completions

Body Parameters

Name	Type	Required	Description
model	string	Yes	Format: gpt-4-gizmo-(id).
messages	array<object>	Yes	Conversation messages.
max_tokens	number	No	Maximum output tokens.
temperature	number	No	Sampling temperature.
stream	boolean	No	Streaming switch.

Body demo

{
  "model": "gpt-4-gizmo-g-bo0FiWLY7",
  "messages": [
    {
      "role": "user",
      "content": "What is quantum computing potential?"
    }
  ],
  "max_tokens": 512,
  "stream": false
}