Back to DeepSeek V3.2 in the gallery

DeepSeek

DeepSeek's successor keeps the V3 template but adds sparse attention to cut long-context costs.

DeepSeek V3.2 architecture diagram
DeepSeek V3.2 architecture diagram. Select the diagram to view it at full size.

Architecture and specifications

Scale
671B total, 37B active (5.5% active)
Context (tokens)
128,000
License
MIT License
Date
2025-12-01
Decoder type
Sparse MoE
Attention
MLA with DeepSeek Sparse Attention
Layer mix
61 MLA/DSA
KV cache / token (bf16) info
83.875 KiB · Moderate
Key detail
Adds sparse attention while retaining the DeepSeek V3-style MTP path.

Training and benchmarks

Pretraining data
943.7B continued-training tokens + 2.1B indexer warm-up Additional DSA training from V3.1-Terminus. The report does not give a cumulative pretraining total. DeepSeek V3.2 report
Post-training methods
SFT, RLVR, RL, Distillation Specialist distillation and mixed GRPO; rule-based reasoning/agent rewards and rubric-based generative rewards. Tool-use cold start uses SFT. DeepSeek V3.2 report
AA Intelligence Index v4.3.2 info
Total score 16.0 General N/A Scientific 6.0 Coding N/A Agents N/A

Further reading and tools