unsloth / image-text-to-text updated 4 months ago

Qwen3.6-35B-A3B-MTP-GGUF

Name: Qwen3.6-35B-A3B-MTP-GGUF
Author: unsloth

See Unsloth Dynamic 2.0 GGUFs for our quantization benchmarks. MTP enables ~1.5-2x faster inference with no accuracy loss. You can now run Qwen3.6 MTP GGUFs in Unsloth Studio Unsloth Studio auto sets the ideal MTP settings for your hardware (Mac, CPU, GPU):

Params

—

Context

—

Downloads 30d

1.0 M

Likes

933

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

Download history

daily snapshots · 9 days

1.0 M840 K

Sep 12Sep 15Sep 18Sep 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
Qwen3.6-35B-A3B-UD-IQ1_M.gguf	IQ1_M	11.4 GB	13.0 GB	✅ Runs comfortably
Qwen3.6-35B-A3B-UD-IQ2_XXS.gguf	IQ2_XXS	11.8 GB	13.5 GB	✅ Runs comfortably
Qwen3.6-35B-A3B-UD-IQ2_M.gguf	IQ2_M	11.9 GB	13.6 GB	✅ Runs comfortably
Qwen3.6-35B-A3B-UD-IQ3_XXS.gguf	IQ3_XXS	14.1 GB	16.0 GB	✅ Runs comfortably
Qwen3.6-35B-A3B-UD-IQ3_S.gguf	IQ3_S	15.3 GB	17.4 GB	✅ Runs comfortably
Qwen3.6-35B-A3B-UD-IQ4_NL.gguf	IQ4_NL	18.5 GB	20.9 GB	⚠️ Tight — reduce context
Qwen3.6-35B-A3B-Q8_0.gguf	Q8_0	37.8 GB	42.1 GB	❌ Won’t fit
Qwen3.6-35B-A3B-BF16-00001-of-00002.gguf	GGUF	49.9 GB	55.4 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

Licence: apache-2.0
First seen on the Hub: 2026-05-11
Base model: Qwen3.6-35B-A3B
Training datasets: undisclosed
Added to our catalog: 2026-09-12

Family

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

Base Qwen3.6-35B-A3B Qwen 3.3 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 Qwen3.5-9B 9.3 M gemma-4-31B-it 9.1 M Qwen3.8-27B 7.4 M

Appears in

Fastest-growing VLMs