Qwen/Qwen2.5-Coder-3B
Full fine-tune vs Qwen/Qwen2.5-3B. 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
34 derivatives checked
- Qwen/Qwen2.5-Coder-3B-Instruct: Full fine-tune
- bartowski/Qwen2.5-Coder-3B-GGUF: Same weights, converted
- WeiboAI/VibeThinker-3B: Full fine-tune
- ThomasBaruzier/Qwen2.5-Coder-3B-Instruct-GGUF: Full fine-tune
- saidutta69/Qwen2.5-Coder-3B-Instruct-heretic: Full fine-tune
- ggml-org/Qwen2.5-Coder-3B-Q8_0-GGUF: Quantization
- unsloth/Qwen2.5-Coder-3B-bnb-4bit: Quantization
- mradermacher/Qwen2.5-Coder-3B-i1-GGUF: Quantization
- fawazo/qwen2.5-coder-3b-pentest-gguf: LoRA-style
- lmstudio-community/Qwen2.5-Coder-3B-GGUF: Quantization
- saidutta69/VibeThinker-3B-heretic: Full fine-tune
- QuantFactory/Qwen2.5-Coder-3B-Instruct-GGUF: Full fine-tune
- unsloth/Qwen2.5-Coder-3B: Identical copy
- mradermacher/Qwen2.5-Coder-3B-GGUF: Same weights, converted
- mlx-community/Qwen2.5-Coder-3B-4bit: Quantization
- ICEPVP8977/Uncensored_Qwen2.5_Coder_3B: LoRA-style
- mlx-community/Qwen2.5-Coder-3B-8bit: Quantization
- Volko76/Qwen2.5-Coder-3B-Instruct-GGUF: Full fine-tune
- MaziyarPanahi/Qwen2.5-Coder-3B-GGUF: Same weights, converted
- zkxxxx/VibeThinker-3B-heretic: Full fine-tune
- MBGMX/Qwen2.5-Coder-3B-Instruct: Full fine-tune
- chetan272006/Qwen2.5-Coder-3B-Instruct: Full fine-tune
- Shrijanagain/TINY-THINKER-INSTRUCT: Full fine-tune
- prithivMLmods/Qwen2.5-Coder-3B-GGUF: LoRA-style
- DevQuasar-5/Qwen.Qwen2.5-Coder-3B-GGUF: Quantization
- arun11karthik/cellsense-fim-3b: Full fine-tune
- RecursiveMAS/Mixture-Code-Qwen2.5-Coder-3B: Full fine-tune
- Nellyw888/VeriReason-Qwen2.5-3b-RTLCoder-Verilog-GRPO-reasoning-tb: Full fine-tune
- LovelyBuggies/2xQwen2.5-Coder-3B-Minister-Main: Full fine-tune
- LovelyBuggies/2xQwen2.5-Coder-3B-Minister-Aux: Full fine-tune
3 other repos hold byte-for-byte identical weight files.