AITRADER/Jackrong-Qwen3.5-9B-Claude-Reasoning-abliterated-mxfp8-MLX

Jackrong Qwen3.5-9B Claude Reasoning - Abliterated (MXFP8 MLX)

This is an abliterated (uncensored), MXFP8 quantized version of Jackrong/Qwen3.5-9B-Claude-4.6-Opus-Reasoning-Distilled, converted to MLX format for Apple Silicon.

What is this model?

The base model is Qwen3.5-9B fine-tuned on Claude 4.6 Opus reasoning data using Unsloth/LoRA. This version is abliterated (safety refusals removed) and quantized to MXFP8 for reduced memory usage (~8.6 GB vs ~16.6 GB bf16).

Abliteration Details

• Method: lukey03 (1 direction, norm-preserving, 3 refinement passes)
• Post-processing: LoRA compliance fine-tuning (rank=64, 80 iterations)
• Quantization: MXFP8 (8.25 bits per weight)
• Format: MLX (Apple Silicon native)
• Tool: OBLITERATUS

Usage

pip install mlx-lm

mlx_lm.generate --model AITRADER/Jackrong-Qwen3.5-9B-Claude-Reasoning-abliterated-mxfp8-MLX --prompt "Explain quantum computing"

Full Precision Version

The bf16 version is available at: AITRADER/Jackrong-Qwen3.5-9B-Claude-Reasoning-abliterated-fp16-MLX

Disclaimer

This model is provided for research purposes. Users are responsible for ensuring their use complies with applicable laws and regulations.