Qwen/Qwen2.5-Coder-32B-Instruct
Full fine-tune vs Qwen/Qwen2.5-Coder-32B. Fine-tune: 100% of the weights changed, across 64 of 64 layers.
Tensors changed: Embeddings 1/1, Norms 129/129, MLP 192/192, Attention 448/448, lm_head 1/1
89 derivatives checked
- Qwen/Qwen2.5-Coder-32B-Instruct-AWQ: Quantization
- lmstudio-community/Qwen2.5-Coder-32B-Instruct-MLX-4bit: Quantization
- lmstudio-community/Qwen2.5-Coder-32B-Instruct-MLX-8bit: Quantization
- bartowski/Qwen2.5-Coder-32B-Instruct-GGUF: Quantization
- Qwen/Qwen2.5-Coder-32B-Instruct-GGUF: Full fine-tune
- unsloth/Qwen2.5-Coder-32B-Instruct-bnb-4bit: Quantization
- lmstudio-community/Qwen2.5-Coder-32B-Instruct-GGUF: Quantization
- unsloth/Qwen2.5-Coder-32B-Instruct: Identical copy
- CISCai/Qwen2.5-Coder-32B-Instruct-SOTA-GGUF: Same weights, converted
- unsloth/Qwen2.5-Coder-32B-Instruct-128K-GGUF: Quantization
- unsloth/Qwen2.5-Coder-32B-Instruct-GGUF: Quantization
- mlx-community/Qwen2.5-Coder-32B-Instruct-4bit: Quantization
- EasierAI/Qwen-2.5-Coder-32B: Full fine-tune
- mlx-community/Qwen2.5-Coder-32B-Instruct-8bit: Quantization
- duydq12/Qwen2.5-Coder-32B-Instruct-FP8-dynamic: Quantization
- Qwen/Qwen2.5-Coder-32B-Instruct-GPTQ-Int4: Quantization
- mradermacher/Qwen2.5-Coder-32B-Instruct-i1-GGUF: Quantization
- Mungert/OlympicCoder-32B-GGUF: Full fine-tune
- Thaurock/Qwen2.5-Coder-32B-abliterated-GGUF: Abliterated
- Mungert/SWE-agent-LM-32B-GGUF: Full fine-tune
- ApolloRaines/Qwen2.5-Coder-32B-Instruct-Jbliterated: Full fine-tune
- tzervas/qwen2.5-coder-32b-bitnet-1.58b: Quantization
- second-state/Qwen2.5-Coder-32B-Instruct-GGUF: Same weights, converted
- featherless-ai-quants/Qwen-Qwen2.5-Coder-32B-Instruct-GGUF: Quantization
- TobiasLogic/Qwen2.5-Coder-32B-abliterated: Abliterated
- 3MPER0RR/Qwen2.5-Coder-32B-Instruct-3MPER0RR-abliterated: Abliterated
- Mungert/SWE-Dev-32B-GGUF: Full fine-tune
- Qwen/Qwen2.5-Coder-32B-Instruct-GPTQ-Int8: Quantization
- drawais/Qwen2.5-Coder-32B-Instruct-NVFP4: Quantization
- mlx-community/Qwen2.5-Coder-32B-Instruct-bf16: Same weights, repacked
34 other repos hold byte-for-byte identical weight files.