Can I run DeepSeek V4 Flash 0731 locally?
DeepSeek V4 Flash 0731 by deepseek-ai has 304B parameters (27B active per token, 256 experts, 6 used per token). At 4-bit with an 8K context it needs about 175 GB of GPU memory. It fits on 6 of 65 GPUs and Macs we track.
GGUF Q8_0 file was 3% of the size expected from the parameter count, so this page uses size estimates instead of real file sizes.
This model uses a newer compressed or sparse attention design. Our KV-cache numbers use the standard formula, so treat long-context figures as an upper bound.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | rough | 107 GB | 108 GB | 110 GB | 119 GB |
| 3-bit (Q3_K_M) | usable | 139 GB | 140 GB | 142 GB | 151 GB |
| 4-bit (Q4_K_M) | good, most popular | 174 GB | 175 GB | 177 GB | 186 GB |
| 5-bit (Q5_K_M) | very good | 203 GB | 203 GB | 206 GB | 215 GB |
| 6-bit (Q6_K) | near lossless | 233 GB | 234 GB | 236 GB | 245 GB |
| 8-bit (Q8_0) | effectively lossless | 302 GB | 302 GB | 305 GB | 314 GB |
| 16-bit (F16/BF16) | original weights | 568 GB | 568 GB | 570 GB | 579 GB |
Which GPUs and Macs run DeepSeek V4 Flash 0731?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Prefer an API?
DeepSeek: DeepSeek V4 Flash 0731 is on OpenRouter at $0.06 per 1M input tokens and $0.18 per 1M output tokens. Price details and history →
Questions
How much VRAM does DeepSeek V4 Flash 0731 need?
About 175 GB at 4-bit (Q4_K_M) with an 8K context: 173 GB of weights (estimated), 0.8 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 302 GB.
What is the smallest GPU that runs DeepSeek V4 Flash 0731?
No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the Mac M3 Ultra 512GB (512 GB).
Can I run DeepSeek V4 Flash 0731 on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M3 Ultra 512GB or larger, using the default macOS GPU memory limit.
Similar models
Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.
DeepSeek V4 Flash DSpark
304B · 175 GB at 4-bitDeepSeek V4 Flash Vision Exp
305B · 175 GB at 4-bitDeepSeek V4 Flash
291B · 167 GB at 4-bitHy3
299B · 173 GB at 4-bitHy3 preview
299B · 173 GB at 4-bitMiMo V2 Flash
310B · 176 GB at 4-bitModel facts
| Hugging Face repo | deepseek-ai/DeepSeek-V4-Flash-0731 |
| Parameters | 304,180,418,494 (304B) · ~27B active (computed from config) |
| Layers / hidden size | 43 / 4096 |
| Attention | GQA; 64 heads, 1 KV heads, head dim 512 |
| KV cache per token (FP16) | 96.8 KiB |
| Max context | 1,048,576 tokens |
| License | mit |
| Released on HF | Jul 31, 2026 |
| Downloads (30d) / likes | 2,200,290 / 4,109 |
Sources: Hugging Face model API · config.json (deepseek-ai/DeepSeek-V4-Flash-0731) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.