DeepSeek V4-Flash

Retired

DeepSeekOpen source

Efficient V4 companion model with 284B total / 13B active parameters and the same one-million-token context window. RETIRED 2026-09-10: superseded by DeepSeek-V4.1-Flash. For compatibility the `deepseek-v4-flash` API id temporarily routes to V4.1-Flash.

Official page ↗Model card ↗Paper ↗

Specifications

License: Open source · MIT
Weights: Downloadable
Architecture: Mixture-of-Experts
Parameters: 284B · 13B 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.