curl --request POST \
  --url https://proxy.training.coreweave.com/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "router",
  "messages": [
    {
      "role": "user",
      "content": "Classify this support request."
    }
  ],
  "metadata": {
    "forge.project_version": "2",
    "gen_ai.conversation.id": "conversation-4827"
  }
}
'

{
  "id": "<string>",
  "object": "<string>",
  "created": 0,
  "model": "<string>",
  "choices": [
    {
      "index": 0,
      "message": {},
      "finish_reason": "<string>"
    }
  ],
  "usage": {}
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

Create a routed chat completion

Send OpenAI-Project: entity/project to select a Model Distillation project, and metadata["forge.project_version"] to select a version (omit for v1). Omit model, leave it blank, or use router for configured routing. Set model to provider/model to override the model while retaining project attribution, conversation trace sampling, and opt-in caching. Override requests use the registered provider and caller parameters, not weighted target parameter defaults or overrides. Invalid headers are rejected without fallback.

POST

chat

completions

curl --request POST \
  --url https://proxy.training.coreweave.com/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "router",
  "messages": [
    {
      "role": "user",
      "content": "Classify this support request."
    }
  ],
  "metadata": {
    "forge.project_version": "2",
    "gen_ai.conversation.id": "conversation-4827"
  }
}
'

{
  "id": "<string>",
  "object": "<string>",
  "created": 0,
  "model": "<string>",
  "choices": [
    {
      "index": 0,
      "message": {},
      "finish_reason": "<string>"
    }
  ],
  "usage": {}
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

{
  "error": {
    "message": "<string>",
    "type": "proxy_error"
  }
}

Authorizations

Authorization

string

header

required

W&B API key whose personal/default entity or teams own the selected task and providers.

Headers

OpenAI-Project

string

Weights & Biases entity/project owning the routing configuration. The caller must have access to the entity. Unknown projects return 404. The validated project is forwarded to Serverless Inference for attribution.

Example:

"my-team/my-project"

Forge-Cache-Mode

enum<string>

Opt a non-streaming project request into exact-response caching. Cache identity includes the project version, the routing revision, the effective upstream request body, and forwarded path/query parameters. Model overrides are additionally isolated by provider configuration, model, and passthrough user identity. readWrite reads and populates the cache, readOnly reads without populating it, and writeOnly populates without reading. Only successful responses up to 8 MiB are stored. Invalid values return HTTP 400. Response caching is not supported for streaming requests.

Available options:

readWrite,

readOnly,

writeOnly

Body

application/json

messages

object[]

required

Minimum array length: 1

Show child attributes

model

string

With OpenAI-Project, omit, leave blank, or use router for configured routing. Use {provider}/{model-id} to override the model within the project.

metadata

object

Show child attributes

stream

boolean

default:false

temperature

number

tools

object[]

Show child attributes

tool_choice

any

response_format

object

Show child attributes

reasoning_effort

string

max_tokens

integer

Required range: 1 <= x <= 9007199254740991

max_completion_tokens

integer

Required range: 1 <= x <= 9007199254740991

{key}

any

Response

Provider response. Streaming requests return server-sent events and end with [DONE].

string

object

string

created

integer

Required range: -9007199254740991 <= x <= 9007199254740991

model

string

choices

object[]

Show child attributes

usage

object

Show child attributes

{key}

any

Last modified on September 30, 2026