Qwen/QwQ-32B-Preview
Full fine-tune vs Qwen/Qwen2.5-32B-Instruct. 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
28 derivatives checked
- bartowski/QwQ-32B-Preview-GGUF: Same weights, converted
- mradermacher/QwQ-32B-Preview-i1-GGUF: Quantization
- mmnga/QwQ-32B-Preview-gguf: Quantization
- unsloth/QwQ-32B-Preview-bnb-4bit: Quantization
- unsloth/QwQ-32B-Preview-GGUF: Quantization
- second-state/QwQ-32B-Preview-GGUF: Same weights, converted
- divpasta123/QwQ-32B-Preview-GGUF: Same weights, converted
- featherless-ai-quants/Qwen-QwQ-32B-Preview-GGUF: Quantization
- gaianet/QwQ-32B-Preview-GGUF: Same weights, converted
- mradermacher/QwQ-32B-Preview-GGUF: Quantization
- DevQuasar-5/Qwen.QwQ-32B-Preview-GGUF: Quantization
- huihui-ai/QwQ-32B-Preview-abliterated: Abliterated
- Saxo/Linkbricks-Horizon-AI-Japanese-Base-32B: LoRA-style
- prithivMLmods/Qwen2.5-32B-DeepSeek-R1-Instruct: Full fine-tune
- karakuri-ai/karakuri-lm-32b-thinking-2501-exp: Full fine-tune
- OpenBuddy/openbuddy-qwq-32b-v24.1-200k: Full fine-tune
- ehristoforu/ud-14b: Full fine-tune
- Sakalti/oxyge1-33B: Full fine-tune
- OpenBuddy/openbuddy-qwq-32b-v24.2-200k: Full fine-tune
- prithivMLmods/Blaze.1-32B-Instruct: LoRA-style
- 6cf/QwQ-32B-Preview-IdeaWhiz-v1: LoRA-style
- YOYO-AI/QwQ-Coder-instruct: Full fine-tune
- pipihand01/QwQ-32B-Preview-abliterated-linear75: Abliterated
- impactframes/QWQCoder: Full fine-tune (fewer layers)
- mergekit-community/QwQ-32B-Preview-Instruct-Coder: Full fine-tune
- pipihand01/QwQ-32B-Preview-abliterated-linear25: Abliterated
- BenevolenceMessiah/QwQ-32B-Preview-Self-instruct-2x-TIES-v1.0: Full fine-tune
- pipihand01/QwQ-32B-Preview-abliterated-linear50: Abliterated