deepseek-ai / text-generation updated 1 year ago

DeepSeek-V3

Name: DeepSeek-V3
Author: deepseek-ai

We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets...

Params

684.5 B

Context

163,840

Downloads 30d

1.2 M

Likes

4,169

Licence: unknown Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 10 days

1.2 M1.2 M

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	f8_e4m3	688.6 GB	860.6 GB	❌ Won’t fit
model.safetensors (bf16, full)	bf16 + 160K ctx	688.6 GB	2,811.5 GB	❌ Won’t fit

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/deepseek-v3
{
  "hf_id": "deepseek-ai/DeepSeek-V3",
  "params_b": 684.53,
  "context_length": 163840,
  "license": { "id": "", "commercial": "unknown" },
  "downloads_30d": 1211587,
  "vram_estimates": [
    { "quant": "f8_e4m3", "gb": 860.6 }
  ],
  "updated_at": "2026-07-28T18:03:33Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: DeepseekV3ForCausalLM
Parameters: 684.5 B
Tensor type: F8_E4M3
Context length: 163,840
Vocabulary: 129,280
Layers / heads: 61 / 128
First seen on the Hub: 2024-12-25
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with

DeepSeek-R1 9.9 M GLM-5.2-FP8 2.7 M GLM-5.2 2.2 M GLM-5.2-NVFP4 1.8 M DeepSeek-V3.2 1.2 M GLM-5-FP8 1.2 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB