Thali

← All models

DeepSeek R1

deepseek/deepseek-r1

available chat hosted: global zero_retention

DeepSeek's reasoning model: emits explicit chain-of-thought and excels at maths, code and multi-step problems at a price far below frontier reasoning models.

Context window 64,000 tokens
Input ₹437.00 per million tokens
Output ₹437.00 per million tokens
Licence varies verify before production use

Call this model

from openai import OpenAI client = OpenAI( base_url="https://thaliai.in/api/v1", api_key="thali-sk-...", ) completion = client.chat.completions.create( model="deepseek/deepseek-r1", messages=[{"role": "user", "content": "Hello"}], )

Works with any OpenAI SDK. Streaming, fallback lists and provider preferences are documented in the routing guide.

This model is served via global infrastructure. For workloads that must stay in India, filter the catalog for hosted_in: "in".