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 |
bigcode
|
|---|---|
| Company | bigcode |
| Release Date | 2023-05-04 |
| Parameters (B) | 15.5 |
| Architecture | Decoder-only Transformer |
| Context Window | 8192 |
| Input Modalities | text |
| Open Source | Yes |
| License | bigcode-openrail-m |
| Score | 0 |
| VRAM (GB) | 31 |
| Compute | Single NVIDIA A100 40GB for bf16 inference (~31 GB VRAM); INT8 quantization possible on ~20 GB GPU |
| Benchmark: GSM8K | 8.4 |
| Benchmark: HumanEval | 30.4 |
| Benchmark: MATH | 15.1 |
| Benchmark: MBPP | 49 |
| Benchmark: MMLU | 34.2 |
| Pricing (Input) | — |
| Pricing (Output) | — |