T5 models
3 T5 models from google-t5 on Hugging Face, from 223M to 2.9B parameters, published by google-t5 in F32, with a 512-token context. The smallest official model, t5-3b, needs about 12.7 GB of VRAM at its published precision; the cheapest live fit is V100 at $0.110/hr.
Pick a size
One row per official T5 size: the VRAM it needs at each precision, the cheapest GPU that holds it today, and the KV cache for a 32K-token context.
| Model | Parameters | Native VRAM | FP8 VRAM | INT4 VRAM | Live GPU fit (native) | Est. $/hr | KV cache at 32K |
|---|---|---|---|---|---|---|---|
| t5-3b | 2.9B | 12.7 GB | 3.2 GB | 1.6 GB | V100 | $0.110/hr | not published for this architecture |
VRAM is weights times a flat 1.2 overhead; the KV cache is a separate per-model figure at 16-bit, shown where the architecture is published. See the methodology.
Official models (1)
Fine-tunes and community models (2)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| chatgpt_paraphraser_on_T5_base | 223M | F32 | 1.0 GB | RTX 3070 | $0.088/hr |
| T5_Paraphrase_Paws | 223M | F32 | 1.0 GB | RTX 3070 | $0.088/hr |
VRAM is for the precision the model is published in: the weights times a flat 1.2 overhead, with the KV cache not included. 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.