nvidia / text-generation updated 1 year ago

Qwen3-235B-A22B-NVFP4

Name: Qwen3-235B-A22B-NVFP4
Author: nvidia

The NVIDIA Qwen3-235B-A22B FP4 model is the quantized version of Alibaba's Qwen3-235B-A22B model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA Qwen3-235B-A22B FP4 model is quantized with TensorRT Model Optimizer.

Params

132.8 B

Context

40,960

Downloads 30d

173 K

Likes

Commercial use: allowed apache-2.0 Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 10 days

173 K158 K

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	u8	134.1 GB	168.0 GB	❌ Won’t fit
model.safetensors (bf16, full)	bf16 + 40K ctx	134.1 GB	247.6 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/qwen3-235b-a22b-nvfp4
{
  "hf_id": "nvidia/Qwen3-235B-A22B-NVFP4",
  "params_b": 132.81,
  "context_length": 40960,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 173037,
  "vram_estimates": [
    { "quant": "u8", "gb": 168.0 }
  ],
  "updated_at": "2026-07-28T18:09:07Z"
}

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

Specifications

Architecture: Qwen3MoeForCausalLM
Parameters: 132.8 B
Tensor type: U8
Context length: 40,960
Vocabulary: 151,936
Layers / heads: 94 / 64
Licence: apache-2.0
First seen on the Hub: 2025-07-08
Base model: Qwen3-235B-A22B
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Family

Base model and the most-downloaded derivatives in the catalog.

Base Qwen3-235B-A22B Qwen 695 K

Compare with

dolphin-2.9.1-yi-1.5-34b 4.6 M gpt-oss-120b 4.2 M NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 2.8 M DeepSeek-V4-Flash 2.7 M Qwen3-Coder-Next-FP8 2.2 M Qwen-72B 2.2 M

Appears in

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