unsloth / image-text-to-text updated 2 months ago

Qwen3.6-27B-NVFP4

Name: Qwen3.6-27B-NVFP4
Author: unsloth

2.5x faster throughput than other NVFP4 quants. This is an Unsloth NVFP4 quantized checkpoint calibrated on a mixture of our Unsloth dataset + UltraChat dataset. Works on a 24GB VRAM GPU. Benchmarks on 1xB200 128 concurrency.

Params

21.2 B

Context

—

Downloads 30d

1.7 M

Likes

281

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

Download history

daily snapshots · 55 days

▲ 2.2 M in the last 30 days (57.5%)

4.1 M1.7 M

Jul 28Aug 15Sep 2Sep 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	f8_e4m3	23.4 GB	29.4 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: 21.2 B
Tensor type: F8_E4M3
Licence: apache-2.0
First seen on the Hub: 2026-04-23
Base model: Qwen3.6-27B
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Family

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

Base Qwen3.6-27B Qwen 3.6 M

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 gemma-4-31B-it 9.1 M Qwen3.8-27B 7.4 M Qwen3.8-27B-FP8 7.1 M