QW

Huihui Qwen3 VL 30B A3B Instruct Abliterated

QwenGGUF by mradermacherUpdated 2026-10-23
Download on Hugging Face
Parameters30.5B
Context256K
Layers48
Architectureqwen3vlmoe
Smallest quant11.3 GB
Quick answer

Huihui Qwen3 VL 30B A3B Instruct Abliterated needs about 19.9 GB of memory at Q4_K_M (a 18.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 11.3 GB (Q2_K) to 32.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.2 GB.

Huihui Qwen3 VL 30B A3B Instruct Abliterated – 30B vision‑language model

This model is a 30‑billion‑parameter Qwen3‑VL variant with 3 billion active mixture‑of‑expert parameters. It supports vision input and was ablitated to reduce safety filtering. The model is released under Apache‑2.0.

Good at

  • Vision‑language instruction following
  • Large context handling up to 262 144 tokens
  • Research‑grade ablitated outputs

Limits

  • No safety filtering – risk of sensitive content
  • Requires 8 k context for memory estimates
  • Not suitable for production use

Best for

  • Controlled research experiments
  • Image‑guided text generation
  • Testing ablitated safety behavior

Who it’s for. Researchers and developers who need a vision‑language model with reduced safety filtering for controlled experiments. The model is not intended for public or commercial deployment.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K11.3 GB 12.6 GB
Too big Get file →
Q3_K_S13.3 GB 14.6 GB
Too big Get file →
Q3_K_M14.7 GB 16.0 GB
Too big Get file →
Q3_K_L15.9 GB 17.2 GB
Too big Get file →
IQ4_XS16.6 GB 17.9 GB
Too big Get file →
Q4_K_S17.5 GB 18.8 GB
Too big Get file →
Q4_K_M18.6 GB 19.9 GB
Too big Get file →
Q5_K_S21.1 GB 22.4 GB
Too big Get file →
Q5_K_M21.7 GB 23.0 GB
Too big Get file →
Q6_K25.1 GB 26.4 GB
Too big Get file →
Q8_032.5 GB 33.8 GB
Too big Get file →

For 8 GB VRAM use Q2_K (11.26 GB file, ~12.6 GB RAM). For 16 GB VRAM Q4_K_M (18.56 GB file, ~19.9 GB RAM). For 32 GB VRAM Q8_0 (32.48 GB file, ~33.8 GB RAM). Estimates are approximate.

What GPU do you need for Huihui Qwen3 VL 30B A3B Instruct Abliterated?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q3_K_M (14.7 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (21.7 GB file, about 23.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (25.1 GB file, about 26.4 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (32.5 GB file, about 33.8 GB in use).

Which Huihui Qwen3 VL 30B A3B Instruct Abliterated quant should you choose?

Q4_K_M is the usual starting point for Huihui Qwen3 VL 30B A3B Instruct Abliterated: it balances file size and quality well. Here it is a 18.6 GB file.

The Q8_0 file is 32.5 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 11.3 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3vlmoe
Parameters30.5B (approximate, mixture of experts)
Active parameters3B
Layers48
Trained context262,144 tokens
KV cache (fp16)96 KB per token

Frequently asked questions about Huihui Qwen3 VL 30B A3B Instruct Abliterated

How much VRAM does Huihui Qwen3 VL 30B A3B Instruct Abliterated need?

About 19.9 GB at Q4_K_M with an 8K context window. That is the 18.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Huihui Qwen3 VL 30B A3B Instruct Abliterated on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Huihui Qwen3 VL 30B A3B Instruct Abliterated GGUF?

It depends on the quant: Q2_K is 11.3 GB, Q3_K_S is 13.3 GB, Q3_K_M is 14.7 GB, Q3_K_L is 15.9 GB, IQ4_XS is 16.6 GB, Q4_K_S is 17.5 GB and 5 more.

What context length does Huihui Qwen3 VL 30B A3B Instruct Abliterated support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.2 GB at 32K.

Where can I download Huihui Qwen3 VL 30B A3B Instruct Abliterated GGUF?

The files are on Hugging Face in the mradermacher/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Huihui Qwen3 VL 30B A3B Instruct Abliterated use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run this model on 8 GB VRAM?

Use the Q2_K quant; it requires about 12.6 GB of RAM at 8 k context, which is close to 8 GB but may need swapping.

Is this model safe for public use?

No. The safety filtering has been significantly reduced; use only in controlled environments.

What license applies to this model?

Apache‑2.0. The model can be used and modified under that license.

Other GGUFs of Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Qwen models

All Qwen →

Other medium models

All medium models →