Together AI now serves the Anthropic Messages API and the OpenAI Responses API (November 3, 2026)

What you get

On November 3, 2026 Together AI added the Anthropic Messages API at POST /v1/messages and the OpenAI Responses API at POST /v1/responses. An app built on either one can run on Together by changing its API key, base URL and model, and keep using the official Anthropic or OpenAI SDK.

• Supported models: zai-org/GLM-5.3 and zai-org/GLM-5.2 on serverless inference; zai-org/GLM-5.3-Flash and zai-org/GLM-5.2 on dedicated model inference. Other serverless models accept the requests but aren't supported and may return malformed output.
• Messages API: messages.create, messages.stream and messages.count_tokens work. messages.batches and models.list don't (use Together's own batch API and model list instead).
• Responses API: only responses.create works. Together doesn't store responses, so send the full conversation with each request.

Prerequisites

• A Together AI API key in TOGETHER_API_KEY
• The Anthropic SDK (pip install anthropic or npm install @anthropic-ai/sdk) or the OpenAI SDK (pip install openai or npm install openai)

Steps

1. Messages API: point the Anthropic SDK at https://api.together.ai:

import os
import anthropic

client = anthropic.Anthropic(
    api_key=os.environ["TOGETHER_API_KEY"],
    base_url="https://api.together.ai",
)

message = client.messages.create(
    model="zai-org/GLM-5.3",
    max_tokens=4096,
    messages=[{"role": "user", "content": "Hello!"}],
)

print(message.content[0].text)