Merged Model Arxiv:2408.07990 Autotrain compatible Base model:bunnycore/llama-3.2... Base model:bunnycore/llama-3.2... Base model:fuseai/fusechat-lla... Base model:meta-llama/llama-3.... Base model:theprint/rewiz-llam... Chat Conversational Doi:10.57967/hf/4781 En Endpoints compatible Facebook Llama Llama-3 Meta Model-index Moonride Pytorch Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/MoonRide/Llama-3.2-3B-Khelavaster
Llama 3.2 3B Khelavaster Benchmarks
MMLU Pro: 23.57
GPQA: 3.69
MUSR: 5.5
BBH: 22.69
IFEval: 49.25 vs 88 (so35)-44%
MATH Lvl 5: 16.16
LLME Score: 0.32608
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)
Llama 3.2 3B Khelavaster Parameters and Internals
| LLM Name | Llama 3.2 3B Khelavaster |
| Repository 🤗 | https://huggingface.co/MoonRide/Llama-3.2-3B-Khelavaster |
| Base Model(s) | |
| Merged Model | Yes |
| Model Size | 3b |
| Required VRAM | 7.2 GB |
| Updated | 2025-03-18 |
| Maintainer | MoonRide |
| Model Type | llama |
| Model Files | |
| Supported Languages | en |
| Model Architecture | LlamaForCausalLM |
| License | llama3.2 |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.49.0 |
| Tokenizer Class | PreTrainedTokenizer |
| Vocabulary Size | 128256 |
| Torch Data Type | float16 |
Best Alternatives to Llama 3.2 3B Khelavaster
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Llama 3.2 3B Instruct | 128K / 6.5 GB | 2881713 | 1227 |
| Hermes 3 Llama 3.2 3B | 128K / 6.5 GB | 342805 | 147 |
| Llama 3.2 3B | 128K / 6.5 GB | 417733 | 528 |
| DeepSeek R1 Distill Llama 3B | 128K / 6.5 GB | 2418 | 10 |
| Llama 3.2 3B RP Toxic Fuse | 128K / 6.4 GB | 43 | 2 |
| Llama 3.2 3B ToxicKod | 128K / 6.4 GB | 61 | 2 |
| ReasoningCore 3B RE1 V2 | 128K / 6.5 GB | 289 | 0 |
| ...penReasoner Llama 3.2 3B Rs1.0 | 128K / 6.5 GB | 199 | 1 |
| Calme 3.1 Llamaloi 3B | 128K / 10.6 GB | 3021 | 1 |
| Eximius Persona 5B | 128K / 11.6 GB | 443 | 4 |
Note: green Score (e.g. "73.2") means that the model is better than MoonRide/Llama-3.2-3B-Khelavaster .
Rank the Llama 3.2 3B Khelavaster 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? 45231 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227