tencent models
2 tencent models on Hugging Face, from 1.1B to 2.0B parameters. At the precision each one is published in, the smallest needs about 2.5 GB of VRAM (HunyuanOCR, cheapest live fit: RTX 3060) and the largest about 4.4 GB (Youtu-LLM-2B, cheapest live fit: RTX 3060). The cheapest way to run HunyuanOCR is $0.110/hr.
tencent models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| HunyuanOCR | 1.1B | BF16 | 2.5 GB | RTX 3060 | $0.110/hr |
| Youtu-LLM-2B | 2.0B | BF16 | 4.4 GB | RTX 3060 | $0.110/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.
