Qwen/QwQ-32B
Full fine-tune vs Qwen/Qwen2.5-32B. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 1/1, Norms 128/129, MLP 192/192, Attention 448/448, lm_head 1/1
75 derivatives checked
- MaziyarPanahi/QwQ-32B-GGUF: Quantization
- lmstudio-community/QwQ-32B-MLX-8bit: Quantization
- Qwen/QwQ-32B-GGUF: Full fine-tune
- Qwen/QwQ-32B-AWQ: Quantization
- bartowski/Qwen_QwQ-32B-GGUF: Quantization
- Alibaba-NLP/WebDancer-32B: Full fine-tune
- lmstudio-community/QwQ-32B-GGUF: Quantization
- unsloth/QwQ-32B-GGUF: Same weights, converted
- gaianet/QwQ-32B-GGUF: Same weights, converted
- brittlewis12/QwQ-32B-GGUF: Same weights, converted
- Mungert/QwQ-32B-ArliAI-RpR-v4-GGUF: LoRA-style
- doggydon007/Qwen-32B-Syntactic-Exploit-Core-BF16: Full fine-tune
- Mungert/Kevin-32B-GGUF: Full fine-tune
- mmnga/QwQ-32B-gguf: Quantization
- mradermacher/QwQ-32B-i1-GGUF: Quantization
- second-state/QwQ-32B-GGUF: Same weights, converted
- ArliAI/QwQ-32B-ArliAI-RpR-v1-GGUF: Full fine-tune
- RedHatAI/QwQ-32B-FP8-dynamic: Quantization
- mradermacher/QwQ-32B-GGUF: Quantization
- featherless-ai-quants/Qwen-QwQ-32B-GGUF: Quantization
- DevQuasar-9/Qwen.QwQ-32B-GGUF: Same weights, converted
- abeja/ABEJA-QwQ32b-Reasoning-Japanese-v1.0: Full fine-tune
- jkim96/QwQ-32B-DASHQ-Q2-GGUF: Quantization
- AlicanKiraz0/Seneca-Cybersecurity-LLM-x-QwQ-32B-Q4_Medium-Version: Quantization
- DhruvalLabs/QwQ-32B-GGUF: Quantization
- huihui-ai/QwQ-32B-abliterated: Abliterated
- rinna/qwq-bakeneko-32b: Full fine-tune
- mlx-community/QwQ-32B-4bit: Quantization
- tensorblock/QwQ-32B-GGUF: Quantization
- unsloth/QwQ-32B-unsloth-bnb-4bit: Quantization
85 other repos hold byte-for-byte identical weight files.