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 |
OpenGVLab
|
|---|---|
| Company | OpenGVLab |
| Release Date | 2024-07-04 |
| Parameters (B) | 8.1 |
| Architecture | InternViT + Transformer |
| Context Window | 8192 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | mit |
| Score | 0 |
| VRAM (GB) | 16 |
| Compute | Single NVIDIA GPU with 16GB+ VRAM for bf16 inference; 8-bit quantization supported; multi-GPU deployment recommended for high-resolution images |
| Benchmark: BBH | 72.7 |
| Benchmark: C-Eval | 80.1 |
| Benchmark: GSM8K | 75.6 |
| Benchmark: HellaSwag | 94.9 |
| Benchmark: HumanEval | 69.5 |
| Benchmark: MATH | 39.5 |
| Benchmark: MBPP | 58.8 |
| Benchmark: MMLU | 73.2 |
| Benchmark: MMMU | 51.8 |
| Benchmark: MMMU-Pro | 29 |
| Pricing (Input) | — |
| Pricing (Output) | — |