BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

DeepSeek R1 671B

By DeepSeek · China

reasoning moe
Parameters
671B
License
MIT
Context
125k
VRAM (Q4)
400 GB
Released
January 2025

Overview

The reference open reasoning model — a 671B MoE with 37B active, released under MIT. Scores 97.3 on MATH-500, 79.8 on AIME, and 90.8 on MMLU.

When to pick this model

  • You're running a dedicated inference server and need frontier reasoning
  • You want the strongest open math, code, and logic model available
  • You need an MIT-licensed model with no commercial restrictions
  • You're benchmarking against closed frontier models like o1 or o3

VRAM requirements by quantization

QuantizationVRAM required
Q4_K_M (recommended)400 GB
Q5_K_M480 GB
Q8_0720 GB
FP16 (no quantization)1342 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Published benchmark scores

BenchmarkScore
MMLU90.8
GPQA Diamond71.5
MATH-50097.3

Scores published by the model author or aggregated from public leaderboards. Re-measured monthly by our editorial team.

Strengths

  • MIT license — no commercial restrictions
  • Reference open reasoning model
  • MATH-500 score of 97.3
  • R1-0528 update further sharpens reasoning

Limitations

  • 400GB+ in Q4 — server-class hardware required
  • Out of reach for any single-machine local setup
  • Very long reasoning traces drive up latency

Architecture & training

Architecture: MoE (inherited from V3) · Multi-head Latent Attention · auxiliary-loss-free · RL-trained

Training: Distillation + multi-stage RL. R1-0528 update (May 2025).

Verdict

The open reasoning gold standard — if you have the hardware to host it.

Quick start

ollama run deepseek-r1:671b

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Tools

Is DeepSeek R1 671B the right pick for you?

Compute self-hosted ROI → Back to catalog