Merged Model Arxiv:2203.05482 Autotrain compatible Base model:bunnycore/llama-3.2... Base model:bunnycore/llama-3.2... Base model:huggingfacetb/finem... Base model:huihui-ai/llama-3.2... Base model:passing2961/thanos-... Base model:prithivmlmods/codep... Base model:prithivmlmods/llama... Conversational Endpoints compatible Instruct Llama Model-index Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/bunnycore/Smol-Llama-3.2-3B
Smol Llama 3.2 3B Benchmarks
MMLU Pro: 24.76
GPQA: 3.58
MUSR: 3.15
BBH: 23.04
IFEval: 66.79 vs 88 (so35)-24.1%
MATH Lvl 5: 13.52
LLME Score: 0.35685
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Smol Llama 3.2 3B Parameters and Internals
| LLM Name | Smol Llama 3.2 3B |
| Repository 🤗 | https://huggingface.co/bunnycore/Smol-Llama-3.2-3B |
| Base Model(s) | |
| Merged Model | Yes |
| Model Size | 3b |
| Required VRAM | 7.2 GB |
| Updated | 2025-01-05 |
| Maintainer | bunnycore |
| Model Type | llama |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.46.2 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|eot_id|> |
| Vocabulary Size | 128256 |
| Torch Data Type | bfloat16 |
Best Alternatives to Smol Llama 3.2 3B
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Llama 3.2 3B Instruct | 128K / 6.5 GB | 1534881 | 849 |
| Orca Mini V9 5 3B Instruct | 128K / 6.5 GB | 177 | 4 |
| Llama 3.2 3B RP DeepThink | 128K / 7.2 GB | 180 | 1 |
| Llama 3.2 3B Instruct | 128K / 6.5 GB | 207707 | 2 |
| Llama 3.2 3B Instruct | 128K / 6.4 GB | 60151 | 42 |
| SKY Ko Llama3.2 3B Lora Epoch2 | 128K / 6.5 GB | 23 | 0 |
| ...lama 3.2 Rabbit Ko 3B Instruct | 128K / 6.5 GB | 986 | 8 |
| SKY Ko Llama3.2 3B Lora Epoch3 | 128K / 6.5 GB | 12 | 0 |
| Codepy Deepthink 3B | 128K / 6.5 GB | 169 | 8 |
| SKY Ko Llama3.2 3B Lora Epoch1 | 128K / 6.5 GB | 14 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than bunnycore/Smol-Llama-3.2-3B .
Rank the Smol Llama 3.2 3B Capabilities
🆘 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! 🌟
What open-source LLMs or SLMs are you in search of? 40868 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227