Anthropic Messages

POST /v1/messages — compatible with Anthropic's Messages API. This is a real adapter layer, not renamed OpenAI fields: system prompts, content blocks, and stop reasons follow Anthropic's shape.

Using the Anthropic SDK

python

from anthropic import Anthropic

client = Anthropic(
    base_url="https://api-gw.aimasteryedu.in",
    api_key="YOUR_API_KEY",
)
msg = client.messages.create(
    model="varunsh/atlas-mini",
    max_tokens=256,
    messages=[{"role": "user", "content": "Hello"}],
)
print(msg.content[0].text)

Request

json

{
  "model": "varunsh/atlas-mini",
  "max_tokens": 256,
  "system": "You are concise.",
  "messages": [
    { "role": "user", "content": "What is an API gateway?" }
  ]
}

Response

json

{
  "id": "msg_...",
  "type": "message",
  "role": "assistant",
  "model": "varunsh/atlas-mini",
  "content": [{ "type": "text", "text": "..." }],
  "stop_reason": "end_turn",
  "usage": { "input_tokens": 18, "output_tokens": 122 }
}

Streaming

With "stream": true, events follow Anthropic's sequence: message_start, content_block_delta, message_delta, message_stop.