Octavian710/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking-uncensored-heretic-GGUF

Same weights, converted vs llmfan46/Qwen3.5-40B-Claude-4.5-Opus-High-Reasoning-Thinking-uncensored-heretic. Same weights stored in another dtype (BF16 → BF16/F32), plus 72 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/241, MLP 0/288, Attention 0/96

On the Hub · Downloads on Model Pulse