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 |
tiiuae
|
|---|---|
| Company | tiiuae |
| Release Date | 2023-04-26 |
| Parameters (B) | 1 |
| Architecture | Transformer |
| Context Window | 2048 |
| Input Modalities | text |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 2.6 |
| Compute | Inference requires ~3GB VRAM in bfloat16 on a single GPU. Trained on 32× NVIDIA A100 40GB GPUs (AWS P4d) using data parallelism with ZeRO via the Gigatron codebase. |
| Benchmark: ARC | 35.1 |
| Benchmark: GSM8K | 53 |
| Benchmark: HellaSwag | 63.6 |
| Benchmark: MMLU | 25.3 |
| Benchmark: TruthfulQA | 36.0 |
| Pricing (Input) | — |
| Pricing (Output) | — |