Autotrain compatible Base model:quantized:unsloth/l... Base model:unsloth/llama-3.2-3... Conversational Dataset:arcee-ai/evolkit-20k Dataset:cognitivecomputations/... Dataset:kingnish/reasoning-bas... En Endpoints compatible Gguf Instruct Llama Model-index Quantized Region:us Rewiz Safetensors Sft Sharded Tensorflow Theprint Trl Unsloth
Model Card on HF 🤗: https://huggingface.co/theprint/ReWiz-Llama-3.2-3B
ReWiz Llama 3.2 3B Benchmarks
MMLU Pro: 20.97
GPQA: 4.47
MUSR: 6.94
BBH: 19.29
IFEval: 46.49 vs 88 (so35)-47.2%
MATH Lvl 5: 9.74
LLME Score: 0.29165
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)
ReWiz Llama 3.2 3B Parameters and Internals
| LLM Name | ReWiz Llama 3.2 3B |
| Repository 🤗 | https://huggingface.co/theprint/ReWiz-Llama-3.2-3B |
| Base Model(s) | |
| Model Size | 3b |
| Required VRAM | 6.5 GB |
| Updated | 2025-02-05 |
| Maintainer | theprint |
| Model Type | llama |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| GGUF Quantization | Yes |
| Quantization Type | gguf |
| Model Architecture | LlamaForCausalLM |
| License | apache-2.0 |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.45.0 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|finetune_right_pad_id|> |
| Vocabulary Size | 128256 |
| Torch Data Type | bfloat16 |
Best Alternatives to ReWiz Llama 3.2 3B
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Llama 3.2 MEDIT 3B O1 | 128K / 6.5 GB | 834 | 12 |
| Gunny | 128K / 6.5 GB | 154 | 0 |
| Llama 3.2 3B COT | 128K / 6.5 GB | 162 | 1 |
| ...2 3B Instruct Unsloth Bnb 4bit | 128K / 2.4 GB | 28215 | 3 |
| Llama 3.2 3B Instruct 4bit | 128K / 1.8 GB | 902013 | 27 |
| Llama 3.2 3B Instruct Bnb 4bit | 128K / 2.2 GB | 261992 | 17 |
| Orca Mini V9 0 3B Instruct | 128K / 6.5 GB | 131 | 5 |
| Llama32 3B En Emo V2 | 128K / 2.2 GB | 130 | 1 |
| Gladiator Mini Exp 1211 3B | 128K / 6.5 GB | 5 | 0 |
| Komodo Llama 3.2 3B V2 Fp16 | 128K / 6.5 GB | 75 | 5 |
Note: green Score (e.g. "73.2") means that the model is better than theprint/ReWiz-Llama-3.2-3B .
Rank the ReWiz 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? 42577 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227