DeepSeek V3.1 Terminus

DeepSeek V3.1 Terminus is a refined release in the DeepSeek V3 mixture-of-experts line, combining V3's general capability with improved reasoning consistency and tool-use reliability. It supports long context and is designed for agentic workflows that mix structured outputs, code, and natural language. Use it as a cost-efficient frontier-tier model for high-volume backend reasoning and agent loops.

Quick start

Use any OpenAI-compatible client. Just change base_url and your key.

from openai import OpenAI

client = OpenAI(
    api_key="tl-xxxxxxxxxxxxxxxxxxxxxxxx",
    base_url="https://api.thalam.ai/v1",
)

response = client.chat.completions.create(
    model="deepseek/deepseek-v3.1-terminus",
    messages=[
        {"role": "user", "content": "Hello from Thalam!"}
    ],
)

print(response.choices[0].message.content)