LLM EXPLORER 60,828 MODELS INDEXED

MiniMax H3 TextEncoder Qwen3VL 32B Abliterated NVFP4 AWQ by pottokao

By pottokao · 98 downloads
HF Repo ↗ Copy URL

MiniMax H3 TextEncoder Qwen3VL 32B Abliterated NVFP4 AWQ is an open-source language model by pottokao. Features: 32b LLM, VRAM: 15.6GB, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.3.

  8-bit   Abliterated   Awq Base model:quantized:qwen/qwen... Base model:qwen/qwen3-vl-32b-i...   Conversational   Endpoints compatible   Image-text-to-text   Instruct   Minimax-h3   Modelopt   Modelopt fp4   Nvfp4   Quantized   Qwen3 vl   Region:us   Safetensors   Sharded   Tensorflow   Text-encoder   Uncensored   Vllm-omni
Model Card on HF 🤗: https://huggingface.co/pottokao/MiniMax-H3-TextEncoder-Qwen3VL-32B-abliterated-NVFP4-AWQ 

MiniMax H3 TextEncoder Qwen3VL 32B Abliterated NVFP4 AWQ Parameters and Internals

LLM NameMiniMax H3 TextEncoder Qwen3VL 32B Abliterated NVFP4 AWQ
Repository 🤗https://huggingface.co/pottokao/MiniMax-H3-TextEncoder-Qwen3VL-32B-abliterated-NVFP4-AWQ 
Base Model(s)  Qwen3 VL 32B Instruct   Qwen/Qwen3-VL-32B-Instruct
Model Size32b
Required VRAM15.6 GB
Updated2026-09-10
Maintainerpottokao
Model Typeqwen3_vl
Instruction-BasedYes
Model Files  5.0 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   0.6 GB: 4-of-4
AWQ QuantizationYes
Quantization Typeawq
Model ArchitectureQwen3VLForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.5.0
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to MiniMax H3 TextEncoder Qwen3VL 32B Abliterated NVFP4 AWQ

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3 VL 32B Instruct AWQ0K / 20.7 GB54954412
Qwen3 VL 32B Instruct Bnb 4bit0K / 20.5 GB2988466
Qwen3 VL 32B Instruct 8bit0K / 36.1 GB29251
... 32B Instruct Unsloth Bnb 4bit0K / 32.5 GB14452
Qwen3 VL 32B Instruct 4bit0K / 19.6 GB3881
Qwen3 VL 32B Instruct0K / 67 GB2881482223
Qwen3 VL 32B Instruct FP80K / 35.7 GB38061947
Cosmos Reason2 32B0K / 66.4 GB266214
... Vl 32B Minimax H3 Fp8 Comfyui0K / 26.5 GB111
...wen3 VL 32B Abliterated XCloud0K / 67 GB60
Note: green Score (e.g. "73.2") means that the model is better than pottokao/MiniMax-H3-TextEncoder-Qwen3VL-32B-abliterated-NVFP4-AWQ.