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 | 2025-08-25 |
| Parameters (B) | 4.7 |
| Architecture | ViT-MLP-LLM |
| Context Window | 32768 |
| Input Modalities | text, image, video |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 10 |
| Compute | Single NVIDIA GPU with ≥10 GB VRAM for bf16 inference; fits on one A100 40GB/80GB. 8-bit quantization reduces requirement to ~6 GB. |
| Benchmark: AIME-2024 | 72.8 |
| Benchmark: AIME-2025 | 57.6 |
| Benchmark: BBH | 78.6 |
| Benchmark: C-Eval | 71.9 |
| Benchmark: GSM8K | 92 |
| Benchmark: HellaSwag | 88.5 |
| Benchmark: HumanEval | 87.8 |
| Benchmark: MATH | 94.4 |
| Benchmark: MMLU | 73.8 |
| Benchmark: MMMU | 64.3 |
| Pricing (Input) | — |
| Pricing (Output) | — |