LLM Releases
← Catalog

DeepSeek V4-Pro

Preview
DeepSeekFrontierOpen source

Preview-series sparse MoE flagship with a one-million-token context window and 1.6T total / 49B active parameters.

Specifications

License
Open source · MIT
Weights
Downloadable
Architecture
Mixture-of-Experts
Parameters
1.6T · 49B active
Context window
1M tokens
Max output
Knowledge cutoff
Price (in / out, $/M)
Modalities
TextCode

Benchmarks

No benchmark scores recorded yet. Spotted some? Submit a correction.

Vendor-reported figures are claims until independently verified. See methodology.