Qwen/Qwen2.5-72B-Instruct
Full fine-tune vs Qwen/Qwen2.5-72B. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, MLP 240/240, Attention 560/560, Norms 147/161, lm_head 1/1
65 derivatives checked
- Qwen/Qwen2.5-72B-Instruct-AWQ: Quantization
- bartowski/Qwen2.5-72B-Instruct-GGUF: Quantization
- Sao10K/72B-Qwen2.5-Kunou-v1: Full fine-tune
- tiny-random/qwen2.5: Different architecture
- Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4: Quantization
- Nexusflow/Athene-V2-Chat: Full fine-tune
- Qwen/Qwen2.5-72B-Instruct-GGUF: Same weights, converted
- mradermacher/Qwen2.5-72B-Instruct-i1-GGUF: Quantization
- zetasepic/Qwen2.5-72B-Instruct-abliterated-GGUF: Quantization
- unsloth/Qwen2.5-72B-Instruct-bnb-4bit: Quantization
- legraphista/Qwen2.5-72B-Instruct-IMat-GGUF: Same weights, converted
- lmstudio-community/Qwen2.5-72B-Instruct-GGUF: Quantization
- axana-labs/Qwen2.5-72B-Instruct-bnb-4bit: Quantization
- AItonomy/PhAI-IDE-72B: LoRA-style
- mradermacher/Qwen2.5-72B-Instruct-GGUF: Quantization
- mlx-community/Qwen2.5-72B-Instruct-4bit: Quantization
- Qwen/Qwen2.5-72B-Instruct-GPTQ-Int8: Quantization
- second-state/Qwen2.5-72B-Instruct-GGUF: Same weights, converted
- huihui-ai/Qwen2.5-72B-Instruct-abliterated: Abliterated
- featherless-ai-quants/Qwen-Qwen2.5-72B-Instruct-GGUF: Quantization
- gaianet/Qwen2.5-72B-Instruct-GGUF: Same weights, converted
- DevQuasar/Qwen2.5-72B-Instruct-GGUF: Quantization
- mesolitica/Qwen2.5-72B-Instruct-FP8: Quantization
- enfuse/Qwen2.5-72B-Instruct-NVFP4: Quantization
- Outlier-Ai/Outlier-150B-V3.2: Different architecture
- CED6688/magnum-v4-72b-AWQ: Quantization
- ThalisAI/Qwen2.5-72B-Instruct-heretic: Full fine-tune
- tensorblock/Qwen2.5-72B-Instruct-GGUF: Quantization
- indigoskyai/Qwen2.5-72B-Instruct-abliterated: Abliterated
- Infermatic/72B-Qwen2.5-Kunou-v1-FP8-Dynamic: Quantization
16 other repos hold byte-for-byte identical weight files.