Skip to main content
POST
/
inference
/
v1
{path}
Inference Post
curl --request POST \
  --url https://api.example.com/inference/v1{path} \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "messages": [],
  "timeout": 123,
  "temperature": 123,
  "top_p": 123,
  "n": 123,
  "stop": "<string>",
  "max_completion_tokens": 123,
  "max_tokens": 123,
  "modalities": [
    "<unknown>"
  ],
  "presence_penalty": 123,
  "frequency_penalty": 123,
  "stream": true,
  "logit_bias": {},
  "user": "<string>",
  "response_format": {},
  "seed": 123,
  "tools": [
    "<unknown>"
  ],
  "tool_choice": "<string>",
  "logprobs": true,
  "top_logprobs": 123,
  "parallel_tool_calls": true,
  "extra_headers": {},
  "functions": [
    "<unknown>"
  ],
  "function_call": "<string>",
  "api_version": "<string>",
  "prompt": "<string>",
  "template_vars": {},
  "vertex_credentials": "<string>"
}
'
{
  "detail": [
    {
      "loc": [
        "<string>"
      ],
      "msg": "<string>",
      "type": "<string>"
    }
  ]
}

Authorizations

​
Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Path Parameters

​
path
string
required

Body

application/json
​
model
string
required
​
messages
any[]
​
timeout
​
temperature
number | null
​
top_p
number | null
​
n
integer | null
​
stop
​
max_completion_tokens
integer | null
​
max_tokens
integer | null
​
modalities
any[] | null
​
presence_penalty
number | null
​
frequency_penalty
number | null
​
stream
boolean | null
​
logit_bias
Logit Bias · object
​
user
string | null
​
response_format
Response Format · object
​
seed
integer | null
​
tools
any[] | null
​
tool_choice
​
logprobs
boolean | null
​
top_logprobs
integer | null
​
parallel_tool_calls
boolean | null
​
extra_headers
Extra Headers · object
​
functions
any[] | null
​
function_call
string | null
​
api_version
string | null
​
prompt
string | null

Reference to a Weave Prompt object (e.g., 'weave:///entity/project/object/prompt_name:version'). If provided, the messages from this prompt will be prepended to the messages in this request. Template variables in the prompt messages can be substituted using the template_vars parameter.

​
template_vars
Template Vars · object

Dictionary of template variables to substitute in prompt messages. Variables in messages like '{variable_name}' will be replaced with the corresponding values. Applied to both prompt messages (if prompt is provided) and regular messages.

​
vertex_credentials
string | null

JSON string of Vertex AI service account credentials. When provided for vertex_ai models (e.g. vertex_ai/gemini-2.5-pro), used for authentication instead of api_key. Not persisted in trace storage.

Response

Successful Response