Qwen/Qwen2.5-Coder-7B-Instruct
Full fine-tune vs Qwen/Qwen2.5-Coder-7B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 196/196, lm_head 1/1
243 derivatives checked
- Orion-zhen/Qwen2.5-Coder-7B-Instruct-AWQ: Quantization
- Qwen/Qwen2.5-Coder-7B-Instruct-AWQ: Quantization
- Qwen/Qwen2.5-Coder-7B-Instruct-GPTQ-Int4: Quantization
- nomic-ai/nomic-embed-code: Full fine-tune
- Qwen/Qwen2.5-Coder-7B-Instruct-GGUF: Same weights, converted
- unsloth/Qwen2.5-Coder-7B-Instruct: Identical copy
- unsloth/Qwen2.5-Coder-7B-Instruct-bnb-4bit: Quantization
- bartowski/Qwen2.5-Coder-7B-Instruct-GGUF: Same weights, converted
- lmstudio-community/Qwen2.5-Coder-7B-Instruct-GGUF: Quantization
- lmstudio-community/Qwen2.5-Coder-7B-Instruct-MLX-4bit: Quantization
- unsloth/Qwen2.5-Coder-7B-Instruct-128K-GGUF: Same weights, converted
- bartowski/Qwen2.5.1-Coder-7B-Instruct-GGUF: Full fine-tune
- lmstudio-community/Qwen2.5-Coder-7B-Instruct-MLX-8bit: Quantization
- unsloth/Qwen2.5-Coder-7B-Instruct-GGUF: Same weights, converted
- CISCai/Qwen2.5-Coder-7B-Instruct-SOTA-GGUF: Same weights, converted
- legraphista/Qwen2.5-Coder-7B-Instruct-IMat-GGUF: Same weights, converted
- Kondara/Qwen2.5-Coder-7B-Instruct-Q4_K_M-GGUF: Quantization
- saidutta69/Qwen2.5-Coder-7B-Instruct-heretic: Abliterated
- Mungert/UIGEN-T2-7B-GGUF: LoRA-style
- Mungert/NextCoder-7B-GGUF: Full fine-tune
- OBLITERATUS/Qwen2.5-Coder-7B-Instruct-OBLITERATED: Partial fine-tune
- ggml-org/Qwen2.5-Coder-7B-Instruct-Q8_0-GGUF: Quantization
- eamiller34/Qwen2.5-Coder-7B-Swift-Lm: LoRA-style
- Snowflake/Arctic-Text2SQL-R1-7B: Full fine-tune
- DuoNeural/Qwen2.5-Coder-7B-Instruct-GGUF: Same weights, converted
- mradermacher/Qwen2.5-Coder-7B-Instruct-i1-GGUF: Quantization
- Mungert/nomic-embed-code-GGUF: Full fine-tune
- zhuyaoyu/CodeV-R1-RL-Qwen-7B: Full fine-tune
- apto-as/Qwen2.5-Coder-7B-Instruct-Q5_K_M-GGUF: Quantization
- mlx-community/Qwen2.5-Coder-7B-Instruct-MLX: Same weights, repacked
47 other repos hold byte-for-byte identical weight files.