DeepSeek V3.2

DeepSeek V3.2 is the latest iteration of DeepSeek's flagship mixture-of-experts series, designed for production-grade general reasoning, coding, and multilingual conversation. The architecture activates a small fraction of its 685B parameters per token, giving it strong throughput for high-volume workloads. Best fit: backend chat, agent loops, document understanding, and any task that benefits from frontier-class reasoning without specialized fine-tuning.

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-v3.2",
    messages=[
        {"role": "user", "content": "Hello from Thalam!"}
    ],
)

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