NovaSky-AI/Sky-T1-32B-Flash
Full fine-tune vs NovaSky-AI/Sky-T1-32B-Preview. Fine-tune: 95% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 0/1, Norms 121/129, MLP 192/192, Attention 447/448, lm_head 0/1
15 derivatives checked
- bartowski/Sky-T1-32B-Flash-GGUF: Same weights, converted
- mradermacher/Sky-T1-32B-Flash-i1-GGUF: Quantization
- mradermacher/Sky-T1-32B-Flash-GGUF: Quantization
- DevQuasar-8/NovaSky-AI.Sky-T1-32B-Flash-GGUF: Same weights, converted
- sm54/QwQ-DeepSeek-R1-SkyT1-Flash-Lightest-32B: Full fine-tune
- DataSoul/QwS-STV1-32B-merge: Full fine-tune
- sm54/QwQ-DeepSeek-R1-SkyT1-Flash-Light-32B: Full fine-tune
- tomasmcm/sky-t1-coder-32b-flash: Full fine-tune
- sm54/FuseO1-QwQ-SkyT1-Flash-32B: Full fine-tune
- sm54/QwQ-DeepSeek-R1-SkyT1-Flash-Lighter-32B: Full fine-tune
- valoomba/rombo-qwq-r1-flash-32B-preview: Full fine-tune
- DataSoul/Sky-T1-32B-Flash-abliterated-merge1: Full fine-tune
- DataSoul/QSQ-sce-32B-merge: Full fine-tune
- DataSoul/Sky-T1-32B-Flash-abliterated-merge1.618: Full fine-tune
- YOYO-AI/QwQ-Sky-T1-Med-32B: Full fine-tune