DeepSeek V3 671B

Overview

DeepSeek's frontier-open MoE — 671B total, 37B active — with multi-head latent attention and an auxiliary-loss-free balancing scheme. The V3.1-Terminus update relicenses under MIT.

When to pick this model

• You're running server-class inference and want frontier-open performance
• You need a non-reasoning frontier model for general chat and code at scale
• You want the MLA architecture's reduced KV-cache footprint
• You can move to V3.1-Terminus for MIT licensing

VRAM requirements by quantization

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.

Strengths

• Frontier-open performance in chat, code, and general tasks
• MLA cuts KV memory significantly vs standard attention
• V3.1-Terminus available under MIT
• Pretrained on 14.8T tokens

Limitations

• Original V3 uses the restrictive DeepSeek License
• 400GB+ in Q4 — server-class hardware only
• Overkill for most workloads under 10B requests/month

Architecture & training

Architecture: MoE 256 experts, 8 active · MLA · auxiliary-loss-free · FP8 training

Training: 14.8T tokens pre-training. V3.1-Terminus (Sep 2025) re-licensed MIT.

Quick start

ollama run deepseek-v3:671b

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