Model Card on HF 🤗: https://huggingface.co/AGI-0/smartllama3.1-8B-001
Smartllama3.1 8B 001 Benchmarks
MMLU Pro: 27.63
GPQA: 7.49
MUSR: 14.4
BBH: 24.86
IFEval: 35.18 vs 88 (so35)-60%
MATH Lvl 5: 12.99
LLME Score: 0.27596
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)
Smartllama3.1 8B 001 Parameters and Internals
| LLM Name | Smartllama3.1 8B 001 |
| Repository 🤗 | https://huggingface.co/AGI-0/smartllama3.1-8B-001 |
| Model Size | 8b |
| Required VRAM | 16.1 GB |
| Updated | 2024-12-22 |
| Maintainer | AGI-0 |
| Model Type | llama |
| Instruction-Based | Yes |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| License | unknown |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.45.0.dev0 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Vocabulary Size | 128256 |
| Torch Data Type | float16 |
Best Alternatives to Smartllama3.1 8B 001
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...a 3 8B Instruct Gradient 1048K | 1024K / 16.1 GB | 5272 | 682 |
| P | 1024K / 16.1 GB | 119 | 0 |
| 93 | 1024K / 16.1 GB | 84 | 0 |
| 117 | 1024K / 16.1 GB | 33 | 0 |
| Mpasila Viking 8B | 1024K / 16.1 GB | 84 | 0 |
| 93a | 1024K / 16.1 GB | 20 | 0 |
| 16 | 1024K / 16.1 GB | 169 | 0 |
| 126 | 1024K / 16.1 GB | 13 | 0 |
| 136 | 1024K / 16.1 GB | 13 | 0 |
| 133 | 1024K / 16.1 GB | 12 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than AGI-0/smartllama3.1-8B-001 .
Rank the Smartllama3.1 8B 001 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? 45494 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227