LLM EXPLORER 63,407 MODELS INDEXED

Clef Flash Heretic by saidutta69

By saidutta69 · 371 downloads
HF Repo ↗ Copy URL

Clef Flash Heretic is an open-source language model by saidutta69. Features: 9.4b LLM, VRAM: 18.8GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.32.

  Abliterated Base model:cloudflare/clef-fla... Base model:finetune:cloudflare...   Classification   Clef   Cloudflare   Consumer-gpu   Conversational   Decensored   Decision-making   En   Endpoints compatible   Gguf   Heretic   Image-text-to-text   Image-text-to-typed-output   Jev   Local-llm   Multimodal   Post-train   Quantized   Qwen3.5   Qwen3 5   Region:us   Reproducible   Safetensors   Sharded   Structured-output   Systemone   Tensorflow   Uncensored   Vision
Model Card on HF 🤗: https://huggingface.co/saidutta69/clef-flash-heretic 

Clef Flash Heretic Parameters and Internals

LLM NameClef Flash Heretic
Repository 🤗https://huggingface.co/saidutta69/clef-flash-heretic 
Base Model(s)  Clef Flash   Cloudflare/clef-flash
Model Size9.4b
Required VRAM18.8 GB
Updated2026-10-08
Maintainersaidutta69
Model Typeqwen3_5
Model Files  17.9 GB   4.9 GB: 1-of-4   5.0 GB: 2-of-4   5.0 GB: 3-of-4   3.9 GB: 4-of-4
Supported Languagesen
GGUF QuantizationYes
Quantization Typegguf
Model ArchitectureQwen3_5ForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version5.18.0
Tokenizer ClassQwen3_5Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to Clef Flash Heretic

Best Alternatives
Context / RAM
Downloads
Likes
Qwen Image 2.1 PE I2I0K / 3.6 GB550142
Qwen Image 2.1 PE T2I0K / 3.6 GB188543
Qwen Image 2.1 PE T2I MLX0K / 18.8 GB7291
Qwen Image 2.1 PE I2I MLX0K / 18.8 GB6431
Clef Flash NVFP40K / 0.2 GB1521
Clef Flash FP80K / 13.6 GB1091
Qwen Image 2.1 PE I2I Heretic0K / 18.8 GB1621
Qwen Image 2.1 PE T2I FP80K / 13.5 GB1361
Qwen Image 2.1 PE I2I FP80K / 13.5 GB1122
Qwen Image 2.1 PE I2I Heretic0K / 18.8 GB781
Note: green Score (e.g. "73.2") means that the model is better than saidutta69/clef-flash-heretic.