Arxiv:0000.00000 Autotrain compatible Granite-3.1 Granitemoe Language Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/ibm-granite/granite-3.1-3b-a800m-base
Granite 3.1 3B A800m Base Benchmarks
MMLU Pro: 8.81
GPQA: 3.69
MUSR: 1.11
BBH: 11.91
IFEval: 29.96 vs 88 (so35)-66%
MATH Lvl 5: 4
LLME Score: 0.31327
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)
Granite 3.1 3B A800m Base Parameters and Internals
| LLM Name | Granite 3.1 3B A800m Base |
| Repository 🤗 | https://huggingface.co/ibm-granite/granite-3.1-3b-a800m-base |
| Model Size | 3b |
| Required VRAM | 6.6 GB |
| Updated | 2024-12-21 |
| Maintainer | ibm-granite |
| Model Type | granitemoe |
| Model Files | |
| Model Architecture | GraniteMoeForCausalLM |
| License | apache-2.0 |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 4.46.0 |
| Tokenizer Class | GPT2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 49152 |
| Torch Data Type | bfloat16 |
Best Alternatives to Granite 3.1 3B A800m Base
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Granite 3.1 3B A800m Instruct | 128K / 6.6 GB | 292 | 6 |
| Granite 3.0 3B A800m Instruct | 4K / 6.8 GB | 4030 | 16 |
| Granite 3.0 3B A800m Base | 4K / 13.5 GB | 3058 | 4 |
| PowerMoE 3B | 4K / 13.5 GB | 42929 | 11 |
Note: green Score (e.g. "73.2") means that the model is better than ibm-granite/granite-3.1-3b-a800m-base .
Rank the Granite 3.1 3B A800m Base 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? 40013 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217