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 | 2025-08-26 |
| Parameters (B) | 15.1 |
| Architecture | ViT-MLP-LLM |
| Context Window | 32768 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | apache-2.0 |
| Score | 0 |
| VRAM (GB) | 32 |
| Compute | Single NVIDIA A100 40GB (or equivalent) for bf16 inference; supports multi-GPU deployment via device_map and 8-bit quantization on ~24GB GPUs |
| Benchmark: AIME-2024 | 77.4 |
| Benchmark: AIME-2025 | 63.9 |
| Benchmark: BBH | 82.3 |
| Benchmark: C-Eval | 80.3 |
| Benchmark: GSM8K | 95.6 |
| Benchmark: HellaSwag | 94.1 |
| Benchmark: HumanEval | 97 |
| Benchmark: MATH | 93.6 |
| Benchmark: MMLU | 81.5 |
| Benchmark: MMMU | 73.3 |
| Pricing (Input) | — |
| Pricing (Output) | — |