BenchMark'd
No ratings yet
Be the first to review this model
A major update combining the strengths of V3 and R1 into a single hybrid model that switches between thinking and non-thinking modes. One model covers both general-purpose and reasoning-heavy use cases via chat template changes. Represents the unification of DeepSeek's general and reasoning capabilities.
Released
August 21, 2025
Parameters
671B (MoE, 37B active)
Context
128K
Pricing
Open Source
Last updated: March 15, 2026
Benchmark scores may vary based on evaluation methodology and conditions.