LLM Name | SmolLM3 3B 3bit |
Repository ๐ค | https://huggingface.co/mlx-community/SmolLM3-3B-3bit |
Base Model(s) | |
Model Size | 3b |
Required VRAM | 1.4 GB |
Updated | 2025-07-22 |
Maintainer | mlx-community |
Model Type | smollm3 |
Model Files | |
Supported Languages | en fr es it pt zh ar ru |
Quantization Type | 3bit |
Model Architecture | SmolLM3ForCausalLM |
License | apache-2.0 |
Context Length | 65536 |
Model Max Length | 65536 |
Transformers Version | 4.54.0.dev0 |
Tokenizer Class | PreTrainedTokenizerFast |
Padding Token | <|im_end|> |
Vocabulary Size | 128256 |
Torch Data Type | bfloat16 |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
---|---|---|---|
SmolLM3 3B 8bit | 64K / 3.3 GB | 479 | 6 |
SmolLM3 3B 4bit | 64K / 1.7 GB | 467 | 4 |
SmolLM3 3B Unsloth Bnb 4bit | 64K / 2.1 GB | 210 | 2 |
SmolLM3 3B | 64K / 6.2 GB | 282990 | 560 |
SmolLM3 3B Base | 64K / 6.2 GB | 5217 | 109 |
SmolLM3 3B | 64K / 6.2 GB | 4927 | 4 |
SmolLM3 3B Bf16 | 64K / 6.1 GB | 304 | 2 |
SmolLM3 3B Base | 64K / 6.2 GB | 56 | 2 |
๐ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐