DeepSeek V3.2
DeepSeek-V3.2 is a large language model designed to harmonize high computational...
About
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 that reduces training and inference cost while preserving quality in long-context scenarios. A scalable reinforcement learning post-training framework further improves reasoning, with reported performance in the GPT-5 class, and the model has demonstrated gold-me
Details
| Modality | text |
| Context Window | 163.8K tokens |
| Release Date | Dec 1, 2025 |
| API Available | Yes |
| Hosting | self-hosted, api |
| Output Speed | 140 tokens/sec |
| Time to First Token | 500ms |
| Quality Index | 86/100 |
| Coding Index | 85/100 |
| Reasoning Index | 82/100 |
Benchmarks
Resources
Tags
Quick Info
- Organization
- DeepSeek
- Pricing
- Free (self-hosted) / $0.26/1M input tokens (API)
- Free Tier
- Yes
- Popularity
- 61/100
- Updated
- Feb 20, 2026
Also in AI Models
Claude Opus 4.6
Opus 4.6 is Anthropic’s strongest model for coding and long-running professional...
Claude Sonnet 4.6
Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier per...
Gemini 3 Flash
Google DeepMind's latest fast and capable multimodal AI model