RadixArk / image-text-to-text updated 4 weeks ago

Qwen3.8-27B-NVFP4

Name: Qwen3.8-27B-NVFP4
Author: RadixArk

The RadixArk Qwen3.8-27B-NVFP4 model is the quantized version of Qwen/Qwen3.8-27B. The quantization was produced at RadixArk using NVIDIA Model Optimizer, following a mixed NVFP4 W4A4 recipe.

Params

18.2 B

Context

—

Downloads 30d

2.1 M

Likes

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

Download history

daily snapshots · 27 days

2.5 M856 K

Aug 25Sep 3Sep 12Sep 20

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	21.9 GB	27.3 GB	❌ Won’t fit

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

Specifications

Architecture: Qwen3_5ForConditionalGeneration
Parameters: 18.2 B
Tensor type: U8
Licence: apache-2.0
First seen on the Hub: 2026-08-14
Base model: Qwen3.8-27B
Training datasets: undisclosed
Added to our catalog: 2026-08-25

Family

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

Base Qwen3.8-27B Qwen 7.4 M Derivative Qwen3.8-27B-DSpark RadixArk 337 K

Compare with any image-text-to-text model

Popular comparisons

Qwen3-VL-8B-Instruct 19.5 M gemma-4-26B-A4B-it 9.9 M Qwen3.6-35B-A3B-FP8 9.8 M Qwen3.5-9B 9.3 M gemma-4-31B-it 9.1 M Qwen3.8-27B 7.4 M