DeepSeek-V4-Pro

The larger DeepSeek-V4 reasoning model, built around long-context attention, MoE routing, and coupled post-training and serving choices.

DeepSeek-V4-Pro is useful to study because the model page points to a whole stack, not just a bigger checkpoint. Its identity comes from how architecture, post-training, and long-context serving are combined.

At a glance

Family
deepseek
Source type
Research
Released
June 2026
Authors
DeepSeek-AI
Modalities
Text
Parameter count
671 billion total parameters
Context length
1,048,576 tokens
Precision
fp8, fp4

What It Is

DeepSeek-V4-Pro is the larger reasoning-oriented model in the DeepSeek-V4 release. It is a text model with a transformer-style backbone, sparse expert blocks, and an unusually large advertised context length.

Inputs And Outputs

The model takes text tokens and produces text tokens. The practical difference from an ordinary chat model is not modality. It is that the stack is designed to preserve useful work over very long prompts and long reasoning traces.

Architecture

The model keeps a transformer skeleton, but the paper highlights manifold-constrained hyper-connections, compressed sparse attention, heavily compressed attention, DeepSeekMoE routing, MLA-style cache compression, and supporting serving systems rather than one single signature block.

Important Modules

These are the modules worth opening if you want to understand why the V4 stack behaves differently from a plain decoder-only transformer.

Training

The public story emphasizes specialist post-training, on-policy distillation, and FP4 quantization-aware work. Those steps matter because the model is presented as a system meant to survive reasoning load and deployment pressure together.

Practical Notes

The million-token context is the headline, but the page is easiest to read as a coordinated stack. Architecture choices reduce long-context cost, and serving choices keep that cost from collapsing the runtime.

Tags

References

  1. DeepSeek-AI. "DeepSeek-V4 Technical Report." 2026.