Create an Anthropic Messages-compatible response

POST /v1/messages

Create an Anthropic Messages-compatible response

cURL

curl --request POST \
  --url https://api.abliteration.ai/v1/messages \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "<string>",
  "max_tokens": 499995,
  "messages": [
    {
      "content": "<string>"
    }
  ],
  "system": "<string>",
  "stream": false,
  "temperature": 123,
  "top_p": 123,
  "top_k": 123,
  "stop_sequences": [
    "<string>"
  ],
  "tools": [
    {}
  ],
  "tool_choice": {
    "name": "<string>",
    "disable_parallel_tool_use": true
  },
  "thinking": {
    "budget_tokens": 499995
  },
  "output_config": {
    "format": {
      "type": "json_schema",
      "schema": {}
    }
  },
  "context_management": {
    "edits": [
      {
        "type": "clear_thinking_20251015",
        "keep": "all"
      }
    ]
  },
  "metadata": {},
  "policy_ids": [
    "<string>"
  ]
}
'

200

{
  "id": "<string>",
  "content": [
    {}
  ],
  "model": "<string>",
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 1,
    "output_tokens": 1,
    "cache_creation_input_tokens": 1,
    "cache_read_input_tokens": 1
  },
  "type": "message",
  "role": "assistant",
  "stop_sequence": "<string>"
}

Headers

Authorization · string | null

Bearer API key.

x-api-key · string | null

API key alternative for Anthropic-compatible clients. Authorization takes precedence when both headers are present.

Body

application/json

Anthropic Messages-compatible request.

model · string · required

Minimum string length: 1

max_tokens · integer · required

Required range: 1 <= x <= 999990

messages · AnthropicMessage · object[] · required

Minimum array length: 1

Show child attributes

system · string System · object[]

stream · boolean · default: false

temperature · number | null

top_p · number | null

top_k · integer | null

stop_sequences · string[] | null

tools · Tools · object[] | null

tool_choice · AnthropicToolChoice · object | null

Show child attributes

thinking · AnthropicThinking · object | null

Show child attributes

output_config · AnthropicOutputConfig · object | null

Show child attributes

context_management · AnthropicContextManagement · object | null

Supported beta context-management subset.

keep=all does not remove any replayed thinking blocks, so it can be safely consumed at the compatibility boundary and omitted from the vLLM provider payload.

Show child attributes

metadata · Metadata · object | null

service_tier · enum<string> | null

Available options: auto, standard_only

policy_ids · string[]

Required string length: 1 - 120

Response

200

application/json

Successful Response

id · string · required

Pattern: ^msg_

content · Content · object[] · required

model · string · required

stop_reason · enum<string> | null · required

Available options: end_turn, max_tokens, stop_sequence, tool_use

usage · AnthropicUsage · object · required

Show child attributes

type · string · default: message

Allowed value: "message"

role · string · default: assistant

Allowed value: "assistant"

stop_sequence · string | null