Compare models
Put up to 4 models beside each other — token prices, context windows, capabilities and provider, from the same catalogue the model pages read.
- DeepSeek V3.2 (Thinking)DeepSeekRemove
| Attribute | DeepSeek V3.2 (Thinking)deepseek-v3.2-thinking |
|---|---|
| Pricing | |
| Input | $0.19 / 1M |
| Output | $0.275 / 1M |
| Cache Write (5m) | $0.19 / 1M |
| Cache Write (1h) | $0.19 / 1M |
| Cache Read | $0.19 / 1M |
| Web Search | $0 / 1M |
| Context | |
| Max context | 163.8K |
| Max output | N/A |
| Capabilities | |
| Vision | Yes |
| Function Calling | Yes |
| JSON Mode | Yes |
| Streaming | Yes |
| Catalogue | |
| Provider | DeepSeek |
| Category | chat |
| Charge type | Pay As You Go |
| Released | — |
| Description | |
| Summary | DeepSeek-V3.2 is an efficiency-focused large model that combines strong reasoning with reliable tool use. It introduces DeepSeek Sparse Attention to lower compute costs for long contexts while preserving quality, and uses large-scale reinforcement learning to reach GPT-5-class reasoning (including top IMO/IOI results). An agentic task-synthesis pipeline improves how it reasons with tools in interactive settings — and developers can toggle reasoning on or off as needed. |