HuggingFaceTB/SmolLM2-360M-Instruct
Full fine-tune vs HuggingFaceTB/SmolLM2-360M. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128
59 derivatives checked
- HuggingFaceTB/SmolVLM-500M-Instruct: Full fine-tune (fewer layers)
- dcostenco/smollm2-360m-aac-watch: Quantization
- HuggingFaceTB/SmolLM2-360M-Instruct-GGUF: Quantization
- bartowski/SmolLM2-360M-Instruct-GGUF: Same weights, converted
- unsloth/SmolLM2-360M-Instruct-GGUF: Same weights, converted
- unsloth/SmolLM2-360M-Instruct: Same weights, repacked
- mlx-community/SmolVLM2-500M-Video-Instruct-mlx: Full fine-tune (fewer layers)
- amai-gsu/SmolLM2-360M-Instruct-Q4_K_M-GGUF: Quantization
- mradermacher/SmolLM2-360M-Instruct-i1-GGUF: Quantization
- CatQualia/gnarp-m2: LoRA-style
- lmstudio-community/SmolLM2-360M-Instruct-GGUF: Quantization
- Mungert/SmolVLM-500M-Instruct-GGUF: Full fine-tune
- mlx-community/SmolLM2-360M-Instruct: Same weights, repacked
- mlx-community/SmolVLM2-256M-Video-Instruct-mlx: Different architecture
- MaziyarPanahi/SmolLM2-360M-Instruct-GGUF: Same weights, converted
- mradermacher/SmolLM2-360M-Instruct-GGUF: Same weights, converted
- kon172verma/intent-classifier: Not comparable yet
- zariness00/smolk12_dpo: Full fine-tune
- trychoosing/revonance-drafter-smollm2-360m: LoRA-style
- Alcoft/SmolLM2-360M-Instruct-GGUF: Same weights, converted
- unsloth/SmolLM2-360M-Instruct-bnb-4bit: Quantization
- mfuntowicz/SmolLM2-360M-Instruct-Q4_K_M-GGUF: Quantization
- DevQuasar-11/HuggingFaceTB.SmolLM2-360M-Instruct-GGUF: Quantization
- Zynerji/Ektome-SmolLM2-360Mi-PristinelyUncensored: Abliterated
- LeviDeHaan/SecInt-SmolLM2-360M-nginx: LoRA-style
- prithivMLmods/SmolLM2-360M-F32-GGUF: Same weights, converted
- mlx-community/SmolVLM-500M-Instruct-4bit: Quantization
- 25b3nk/smollm2-360m-boolq-calibration-grpo2-v1: Full fine-tune
- 25b3nk/smollm2-360m-boolq-calibration-sft-v2: Full fine-tune
- xv0y5ncu/SmolLM2-360M-Instruct-GLQ-4bpw: Different architecture
3 other repos hold byte-for-byte identical weight files.