SF·SIX / MODEL PROFILE
DeepSeek V3.2
DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...
DeepSeek V3.2 costs $0.2088 per 1M input tokens and $0.3096 per 1M output tokens at the standard tier, the lowest default-region price among the 12 providers SF·SIX tracks, from GMI Cloud (fp8). Cached input reads cost $0.0216 per 1M tokens. Updated ; prices checked .
- Developer
- DeepSeek
- Input modalities
- text
- Output modalities
- text
- Context tokens
- 163,840
- Latest price observation
- 2026-10-17
- Lowest paired standard token quote
- $0.2088 input / $0.3096 output per 1M tokens
Dated price and provider records
| Provider and channel | Input | Output or native rate | Unit and status | Source and observation |
|---|---|---|---|---|
| GMI CloudOpenRouter | $0.2088 | $0.3096 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| AtlasCloudOpenRouter | $0.26 | $0.38 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| DeepInfraOpenRouter | $0.26 | $0.38 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| Venice AIOpenRouter | $0.2683 | $0.3902 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| BaiduOpenRouter | $0.28 | $0.42 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| SiliconFlowOpenRouter | $0.259 | $0.80 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| DigitalOceanOpenRouter | $0.30 | $0.96 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| FriendliOpenRouter | $0.50 | $1.50 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| GoogleOpenRouter | $0.56 | $1.68 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| PhalaOpenRouter | $1.00 | $1.00 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| MaraOpenRouter | $3.00 | $4.50 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
| SambaNovaOpenRouter | $3.00 | $4.50 | 1M tokenspriced · USD | OpenRouter2026-10-17 |
These are dated, non-executable observations. Verify the exact endpoint, price and terms with the source before use. Unknown values are not free.