DeepSeek V4 Flash 0731
DeepSeek V4 Flash 0731 is an open-weights language model from DeepSeek, released July 31, 2026. It has 304B parameters (mixture-of-experts, 34B active). The context window is 1M tokens. On the Epoch Capabilities Index it scores 154.4, rank 51 of 256 rated models.
DeepSeek-V4-Flash-0731 is the official release of DeepSeek-V4-Flash, superseding the preview version, with substantially enhanced agentic capabilities. It has the same model structure as DeepSeek-V4-Flash-DSpark, i.e. it comes with a speculative decoding module attached.
Check it against your own PC →
Can you run DeepSeek V4 Flash 0731 locally?
Even with 32 GB of VRAM and 32 GB of RAM it is too large to run at a usable speed on a single consumer GPU. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Too large | – | – | – | – |
| 8 GB VRAM | Too large | – | – | – | – |
| 12 GB VRAM | Too large | – | – | – | – |
| 16 GB VRAM | Too large | – | – | – | – |
| 24 GB VRAM | Too large | – | – | – | – |
| 32 GB VRAM | Too large | – | – | – | – |
Ready-made quantizations on Hugging Face: huihui-ai/Huihui-DeepSeek-V4-Flash-0731-abliterated-GGUF · unsloth/DeepSeek-V4-Flash-0731-GGUF · lmstudio-community/DeepSeek-V4-Flash-0731-GGUF · KeinNiemand/DeepSeek-V4-Flash-0731-IK_GGUF · Lucebox/DeepSeek-V4-Flash-0731-ROCmFP3 · AtomicChat/DeepSeek-V4-Flash-0731-GGUF
Strengths by category
ModelTape's 0–100 fit on Epoch AI benchmark results.
| Category | Score | Based on |
|---|---|---|
| Reasoning | 19.9 | Chess Puzzles, SimpleBench, Mystery Game Puzzles |
| Math | 50.6 | ProofBench, OTIS Mock AIME 2024-2025, FrontierMath-Tiers-1-3-v2-Private, FrontierMath-Tier-4-v2-Private |
| Knowledge | 71.4 | GPQA diamond, SimpleQA Verified |
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| OTIS Mock AIME | 94.4% |
| GPQA Diamond | 91.7% |
| FrontierMath (Tiers 1–3) | 60.6% |
| ProofBench | 56.0% |
| SimpleBench | 53.3% |
| SimpleQA Verified | 33.6% |
| Chess Puzzles | 28.5% |
| Mystery Game Puzzles | 27.3% |
| FrontierMath-Tier-4-v2-Private | 24.4% |
Links
Announcement · Hugging Face: deepseek-ai/DeepSeek-V4-Flash-0731 · Epoch AI benchmarks
Models with a similar score
Gemini 3.7 Flash · Muse Spark 1.3 · GPT-5.4 · Grok 4.6 · Qwen 3.8 Max · Claude Opus 4.7 · GPT-5.2 Pro · GLM-5.3
More from DeepSeek
DeepSeek V4.1 Flash 1017 · DeepSeek V4 Flash Vision Exp · DeepSeek V4.1 Flash · DeepSeek V4 Pro 0813 · DeepSeek V4 Pro DSpark · DeepSeek V4 Flash DSpark · DeepSeek-V4-Pro · DeepSeek-V4-Flash