DeepSeek

DeepSeek V4 Flash Flash

304,2B

Parameters

145,6 Go

MXFP4 file

≥ 147,1 Go

Recommended VRAM

1 048 576

Max context

Not measured yet

No card has run the TokenBench protocol on this model yet. The current protocol covers a selection of models; benchmarking any catalog model from the app is coming soon.

Measure my card →

Which cards should it fit on?

Cards on the site whose video memory covers the recommended 147,1 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.

Technical sheet

Published: 2026-07-31
Architecture: 43 layers · 1 KV heads · dimension 512
Attention: Standard: every layer keeps the whole context in memory.
Context cache (f16): ≈ 84 MB per 1,000 tokens · 2,7 Go for 32,768 tokens
GGUF file: bartowski/DeepSeek-V4-Flash-0731-GGUF · 4 files Download (145,6 Go)
Official repo: deepseek-ai/DeepSeek-V4-Flash-0731

Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.