Qwen models
4 Qwen models on Hugging Face, from 782M to 72.3B parameters. At the precision each one is published in, the smallest needs about 1.7 GB of VRAM (nb-asr-beta-qwen06b-lunde05, cheapest live fit: RTX 5060 Ti) and the largest about 162 GB (Qwen-72B, cheapest live fit: RTX A5000). The cheapest way to run nb-asr-beta-qwen06b-lunde05 is $0.110/hr.
Qwen models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| nb-asr-beta-qwen06b-lunde05 | 782M | BF16 | 1.7 GB | RTX 5060 Ti | $0.110/hr |
| Qwen-7B | 7.7B | BF16 | 17.3 GB | RTX A5000 | $0.176/hr |
| Qwen-7B-Chat | 7.7B | BF16 | 17.3 GB | RTX A5000 | $0.176/hr |
| Qwen-72B | 72.3B | BF16 | 162 GB | RTX A5000 × 7 | $1.23/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.
