InternVL2 models
4 InternVL2 models on Hugging Face, from 938M to 25.5B parameters. At the precision each one is published in, the smallest needs about 2.1 GB of VRAM (InternVL2-1B, cheapest live fit: RTX 5060 Ti) and the largest about 57.0 GB (InternVL2-26B, cheapest live fit: A100). The cheapest way to run InternVL2-1B is $0.110/hr.
InternVL2 models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| InternVL2-1B | 938M | BF16 | 2.1 GB | RTX 5060 Ti | $0.110/hr |
| InternVL2-2B | 2.2B | BF16 | 4.9 GB | RTX 5060 Ti | $0.110/hr |
| InternVL2_5-4B | 3.7B | BF16 | 8.3 GB | RTX 5060 Ti | $0.110/hr |
| InternVL2-26B | 25.5B | BF16 | 57.0 GB | A100 | $1.21/hr |
VRAM is for the precision the model is published in, with the same overhead and an 8,192-token context assumed on every page; see the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
More from OpenGVLab
- InternVL3 (4 models)
