Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
ibm-granite
|
|---|---|
| Company | ibm-granite |
| Release Date | 2024-10-21 |
| Parameters (B) | 2.5 |
| Architecture | Transformer |
| Context Window | 4096 |
| Input Modalities | text |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 6 |
| Compute | Consumer GPU with ~6GB VRAM for FP16 inference (2.5B params); runs on CPU with sufficient RAM |
| Benchmark: AGIEval | 29.8 |
| Benchmark: ARC | 54.4 |
| Benchmark: Arena-Elo | 1156 |
| Benchmark: BBH | 43.7 |
| Benchmark: GPQA | 28.6 |
| Benchmark: GSM8K | 59.7 |
| Benchmark: HellaSwag | 76.8 |
| Benchmark: HumanEval | 50.6 |
| Benchmark: IFEval | 46.1 |
| Benchmark: MATH | 23.7 |
| Benchmark: MBPP | 41 |
| Benchmark: MMLU | 56.0 |
| Benchmark: MMLU-Pro | 27.9 |
| Benchmark: TruthfulQA | 53.4 |
| Pricing (Input) | — |
| Pricing (Output) | — |