Back to Models
DeepSeekOpen Source
DeepSeekMath-V2
DeepSeekMath-V2 is a reasoning model developed by DeepSeek-AI. It has a massive parameter count of approximately 6.85 trillion and supports a long context window of 108K.
Parameters
6850.0B
Context Window
108K
License
Apache 2.0
Release Date
2025-11-27
API Pricing
API pricing for this model is not yet available
Strengths
- ・Extremely large number of parameters
- ・Possesses advanced reasoning capabilities
- ・108K long context processing capability
Weaknesses
- ・Requires enormous computational resources
- ・Operational costs due to model size
- ・Possibility of impact on inference speed
Use Cases
- ・Solving complex mathematical problems
- ・Advanced logical reasoning tasks
- ・Large-scale document analysis
Deep Analysis
Base Model
V3.2-Exp-Base (685B MoE)
IMO 2025
5/6 problems (83.3%)
CMO 2024
5/6 problems (73.8%)
Putnam 2024
118/120 (98.3%)
Surpasses human best (90)
License
Apache 2.0
Release Date
November 27, 2025
Strengths
- ・IMO 2025 gold medal (5/6)
- ・Putnam 2024 near-perfect (118/120 vs human best 90)
- ・Self-verifiable reasoning framework
- ・Open-source with full methodology
Weaknesses
- ・Specialized for math theorem proving
- ・High compute for best results
- ・Not a general-purpose model
Competitor Comparison
| Model | GPQA |
|---|---|
| Gemini 2.5 Pro | ~80% |
| GPT-5 Thinking | 85.7% |
DeepSeekMath-V2 achieves gold-medal IMO 2025 (5/6) and near-perfect Putnam 2024 (118/120 vs human best 90). Novel self-verifiable reasoning.
Sources
Analysis generated: 2026-05-24