Qwen/Qwen2.5-Coder-14B
Full fine-tune vs Qwen/Qwen2.5-14B. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 336/336, lm_head 1/1
24 derivatives checked
- Qwen/Qwen2.5-Coder-14B-Instruct: Full fine-tune
- bartowski/Qwen2.5-Coder-14B-GGUF: Same weights, converted
- lmstudio-community/Qwen2.5-Coder-14B-GGUF: Quantization
- unsloth/Qwen2.5-Coder-14B-bnb-4bit: Quantization
- featherless-ai-quants/Qwen-Qwen2.5-Coder-14B-GGUF: Quantization
- QuantFactory/Qwen2.5-Coder-14B-Instruct-GGUF: Full fine-tune
- ggml-org/Qwen2.5-Coder-14B-Q8_0-GGUF: Quantization
- mradermacher/Qwen2.5-Coder-14B-GGUF: Quantization
- DevQuasar-8/Qwen.Qwen2.5-Coder-14B-GGUF: Same weights, converted
- richardyoung/Qwen2.5-Coder-14B-Instruct-heretic: Full fine-tune
- Ahmad-Abdo-Shbat/TIA2.1-14B-GGUF: Full fine-tune
- the-clanker-lover/steelman-14b-ada-GGUF: Full fine-tune
- mlx-community/Qwen2.5-Coder-14B-4bit: Quantization
- unsloth/Qwen2.5-Coder-14B: Identical copy
- mlx-community/Qwen2.5-Coder-14B-bf16: Same weights, repacked
- mlx-community/Qwen2.5-Coder-14B-8bit: Quantization
- MaziyarPanahi/Qwen2.5-Coder-14B-GGUF: Same weights, converted
- itlwas/Qwen2.5-Coder-14B-Q4_K_M-GGUF: Quantization
- RDson/WomboCombo-R1-Coder-14B-Preview: Full fine-tune
- sealad886/FuseO1-DeepSeekR1-Qwen2.5-Coder-14B-Preview: Full fine-tune
- YOYO-AI/Qwen2.5-Coder-14B-YOYO: Full fine-tune
- realYinkaIyiola/YinkaMath-14B: Full fine-tune
- lurf21/Qwen2.5-Coder-14B-NEP: LoRA-style
- Pomoika24/C1-Tachu: LoRA-style
3 other repos hold byte-for-byte identical weight files.