DeepSeek V3.2 vs DeepSeek/DeepSeek-V3.2-Exp-Thinking
DeepSeek V3.2 is cheaper on output tokens, while DeepSeek V3.2 offers a larger context window. Choose DeepSeek V3.2 or DeepSeek/DeepSeek-V3.2-Exp-Thinking based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | DeepSeek V3.2 | DeepSeek/DeepSeek-V3.2-Exp-Thinking |
|---|---|---|
| Provider | DeepSeek | Qiniu |
| Input / 1M tokens | $0.28 | $0.29 |
| Output / 1M tokens | $0.42 | $0.43 |
| Context window | 164K | 128K |
| Parameters | — | — |
| Open weights | — | No |
| Released | — | Sep 2025 |
| LiveBench score | 63.1 | — |
| SWE-bench Multilingual score | 59.0 | — |
| SWE-bench Verified score | 70.0 | — |
| Aider Polyglot score | 74.2 | — |
| LMArena (Chatbot Arena) Elo score | 1425 | — |
| LMArena WebDev Elo score | 1362 | — |
| Epoch Capabilities Index score | 146 | — |
FAQ
Is DeepSeek V3.2 or DeepSeek/DeepSeek-V3.2-Exp-Thinking cheaper?
DeepSeek V3.2 is cheaper on output tokens ($0.42 vs $0.43 per 1M).
Which has the larger context window, DeepSeek V3.2 or DeepSeek/DeepSeek-V3.2-Exp-Thinking?
DeepSeek V3.2 has the larger context window (164K tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.