LOREA-cyber-v5.4

LOREA-cyber v5.4

Qwythos 9B (Qwen3.5) with the v5.4 LoRA merged in, 4-bit MLX. No adapter needed at inference.

Superseded by LOREA-cyber-v5.5. Use that instead. v5.4 is kept here for reference and reproducibility.

Why it was replaced

v5.4 regressed against its own base on the knowledge benchmarks. Two causes, both found by inspecting the data and the training curve afterwards:

• The knowledge MCQs were model-generated and never answer-verified, so some taught the wrong answer. CyberMetric dropped about 5 points.
• It overtrained. Loss fell to 0.37 and general ability went with it, including coding.

v5.5 replaces the synthetic MCQs with real answer-keyed data (CyberMetric, WMDP-cyber, MMLU), trains more gently at 2.5e-5, and selects the checkpoint on full metrics rather than a small subset.

Run it

python3 -m mlx_lm.chat --model MK4-Research/LOREA-cyber-v5.4

Scope

Authorized security testing, CTF and reverse-engineering practice, and education. 4-bit 9B, so verify anything important.