Qwen/Qwen3-0.6B
Full fine-tune vs Qwen/Qwen3-0.6B-Base. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 106/113, MLP 84/84, Attention 112/112
405 derivatives checked
- k2-fsa/OmniVoice: Full fine-tune
- jinaai/jina-reranker-v3: Full fine-tune
- Qwen/Qwen3-0.6B-GGUF: Full fine-tune
- Qwen/Qwen3-0.6B-FP8: Quantization
- MaziyarPanahi/Qwen3-0.6B-GGUF: Same weights, converted
- unsloth/Qwen3-0.6B: Same weights, repacked
- Qwen/Qwen3Guard-Gen-0.6B: Full fine-tune
- unsloth/Qwen3-0.6B-GGUF: Same weights, converted
- Skywork/Skywork-Reward-V2-Qwen3-0.6B: Full fine-tune
- bartowski/Qwen_Qwen3-0.6B-GGUF: Same weights, converted
- mlx-community/Qwen3-0.6B-8bit: Quantization
- unsloth/Qwen3-0.6B-unsloth-bnb-4bit: Quantization
- ggml-org/Qwen3-0.6B-GGUF: Same weights, converted
- mlx-community/Qwen3-0.6B-4bit: Quantization
- jinaai/jina-reranker-v3.5: Full fine-tune
- lmstudio-community/Qwen3-0.6B-MLX-4bit: Quantization
- lmstudio-community/Qwen3-0.6B-MLX-bf16: Same weights, repacked
- C-Tianyu/NanoJev: Partial fine-tune
- charent/self_cognition_Alice: Adapter
- RedHatAI/Qwen3-0.6B-quantized.w4a16: Quantization
- dllm-hub/Qwen3-0.6B-diffusion-mdlm-v0.1: Full fine-tune
- icml2026-7516/ThinkSafe-Qwen3-0.6B: Adapter
- NobodyWho/Qwen_Qwen3-0.6B-GGUF: Quantization
- bd2lcco/Qwen3-0.6B-finetuned: Partial fine-tune
- reaperdoesntknow/Symbiotic-1B: Same weights, converted
- lmstudio-community/Qwen3-0.6B-GGUF: Quantization
- HamoAI/hamo-score-0.6b: LoRA-style
- Qwen/Qwen3Guard-Stream-0.6B: Partial fine-tune
- reaperdoesntknow/Qwen3-0.6B-Distilled-30B-A3B-Thinking-SFT: Partial fine-tune
- geoffmunn/Qwen3-0.6B-f16: Not comparable yet
176 other repos hold byte-for-byte identical weight files.