Qwen/Qwen2.5-Coder-3B-Instruct
Full fine-tune vs Qwen/Qwen2.5-Coder-3B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 73/73, MLP 108/108, Attention 252/252
82 derivatives checked
- Qwen/Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- bartowski/Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- unsloth/Qwen2.5-Coder-3B-Instruct-bnb-4bit: Quantization
- lmstudio-community/Qwen2.5-Coder-3B-Instruct-GGUF: Quantization
- bocalan/Qwen2.5-Coder-3B-Instruct-Q4_K_M-GGUF: Quantization
- unsloth/Qwen2.5-Coder-3B-Instruct: Identical copy
- ryzdfm/qwen2.5-coder-3b-claude_opus_4.6-distilled: LoRA-style
- unsloth/Qwen2.5-Coder-3B-Instruct-128K-GGUF: Same weights, converted
- Qwen/Qwen2.5-Coder-3B-Instruct-AWQ: Quantization
- lmstudio-community/Qwen2.5-Coder-3B-Instruct-MLX-4bit: Quantization
- unsloth/Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- mlx-community/Qwen2.5-Coder-3B-Instruct-4bit: Quantization
- ggml-org/Qwen2.5-Coder-3B-Instruct-Q8_0-GGUF: Quantization
- mradermacher/Qwen2.5-Coder-3B-Instruct-i1-GGUF: Quantization
- lmstudio-community/Qwen2.5-Coder-3B-Instruct-MLX-8bit: Quantization
- k3ssdev/Qwen2.5-Coder-3B-Pentest: LoRA-style
- mradermacher/Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- driaforall/Tiny-Agent-a-3B: Full fine-tune
- tthanhannguyen/qwen25-coder-3b-uncensored: Quantization
- sandeeprdy1729/PrismaCoder-3B: LoRA-style
- second-state/Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- exeterminal/Exe-Guard-Dynamic-GGUF: LoRA-style
- gaianet/Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- Qwen/Qwen2.5-Coder-3B-Instruct-GPTQ-Int4: Quantization
- bekoozkan/godot-qwen2.5-coder-3b-instruct: Adapter
- QuantFactory/Arch-Function-3B-GGUF: Full fine-tune
- DevQuasar-8/Qwen.Qwen2.5-Coder-3B-Instruct-GGUF: Same weights, converted
- Elfsong/Qwen2.5-Coder-3B-Instruct-Venus-Cold-Start: Full fine-tune
- Mungert/Arch-Agent-3B-GGUF: Full fine-tune
- jmarceno/Elixir-Qwen2.5-Coder-3B: LoRA-style
10 other repos hold byte-for-byte identical weight files.