Yi-1.5 models
6 Yi-1.5 models on Hugging Face, from 6.1B to 34.4B parameters. At the precision each one is published in, the smallest needs about 13.5 GB of VRAM (Yi-1.5-6B, cheapest live fit: RTX 5060 Ti) and the largest about 76.9 GB (Yi-1.5-34B-Chat, cheapest live fit: A100). The cheapest way to run Yi-1.5-6B is $0.110/hr.
Yi-1.5 models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| Yi-1.5-6B | 6.1B | BF16 | 13.5 GB | RTX 5060 Ti | $0.110/hr |
| Yi-1.5-9B | 8.8B | BF16 | 19.7 GB | RTX A5000 | $0.176/hr |
| Yi-1.5-9B-Chat | 8.8B | BF16 | 19.7 GB | RTX A5000 | $0.176/hr |
| Yi-1.5-34B-Chat | 34.4B | BF16 | 76.9 GB | A100 | $1.21/hr |
| YiSM-34B-0rn | 34.4B | F16 | 76.9 GB | A100 | $1.21/hr |
| dolphin-2.9.1-yi-1.5-34b | 34.4B | BF16 | 76.9 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 01.AI
- Yi (5 models)
