RUN THIS LLM
Search local LLM hardware requirements
DeepSeek V4 Flash
DeepSeek · 284B MoE · General
284B MoE with 13B active params and 256 experts. Efficiency-focused sibling of V4 Pro — hybrid attention cuts KV cache to ~10% of V3 while keeping a 1M-token context.
VRAM Requirements
| Quantization | VRAM |
|---|---|
| Q4_K_M (smallest) | 170.4 GB |
| Q8_0 (balanced) | 312.4 GB |
| FP16 (full quality) | 568 GB |
Specifications
- Parameters: 284B MoE (13B active per token)
- Category: General
- Max context: 1M tokens
- System RAM: 192 GB minimum
- HuggingFace: deepseek-ai/DeepSeek-V4-Flash
Benchmarks
- MMLU: 88 — General knowledge & reasoning (5-shot)
- HumanEval: 88 — Code generation (pass@1)
Loading interactive analysis...