DeepSeek: DeepSeek V3.2 Thinking
by Kilo Gateway · family deepseek · listed Dec 2025RT
DeepSeek‑V3.2 from DeepSeek harmonizes high computational efficiency with superior reasoning and agent performance. It builds on three main techniques: DeepSeek Sparse Attention for long‑context efficiency, a scalable reinforcement learning framework, and a large‑scale agentic task synthesis pipeline. This model excels at long-context reasoning and agentic tasks, efficiently handling extended inputs while maintaining strong accuracy. Its sparse attention design enables it to process complex, multi-step workflows without excessive compute costs. Overall, DeepSeek‑V3.2 targets long‑context reasoning, tool‑using agents, and efficient deployment in production environments.
Current rates USD per 1M tokens
Against the market
On the record
- Model id
deepseek/deepseek-v3.2-thinking- Context window
- 128K tokens
- Max output
- 8K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- yes
- Open weights
- no
- Knowledge cutoff
- —
- Released
- 2025-12-01
- Last updated
- 2025-12-01
- Provider docs
- kilo.ai
Back of the envelope
≈
Also listed at
- 302.AI$0.29 in · $0.43 out
- Vercel AI Gateway$0.62 in · $1.85 out