anicka/cve-backport-codegen-qwen25-32b

Document safety posture, prompt echo regression, and test generation limits

Add ollama Modelfile

Remove authorship section from model card

Add v4 Q8_0 GGUF (36K dataset + test generation, 93% recall)

v4 model card: 93% recall, test generation, Gemini comparison

Add v3 Q8_0 GGUF (35K dataset, 94% recall, 98% precision)

Modelfile for v3

Add model card with v1/v2/v3 comparison

Copy tokenizer_config.json from v1 repo

Copy tokenizer.json from v1 repo

Copy generation_config.json from v1 repo

Copy config.json from v1 repo

Copy chat_template.jinja from v1 repo

Copy Modelfile from v1 repo

initial commit