pfnet models
3 pfnet models from pfnet on Hugging Face, from 890M to 31.8B parameters, published by pfnet in BF16, with a 256K-token context. The smallest official model, plamo-3-translate-0.8b, needs about 2.0 GB of VRAM at its published precision; the cheapest live fit is RTX 3070 at $0.132/hr.
Pick a size
One row per official pfnet 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 |
|---|---|---|---|---|---|---|---|
| plamo-3-translate-0.8b | 890M | 2.0 GB | 1.0 GB | 0.5 GB | RTX 3070 | $0.132/hr | 0.25 GB |
| plamo-3-translate-8b | 8.5B | 19.1 GB | 9.5 GB | 4.8 GB | RTX 3090 | $0.162/hr | 2.0 GB |
| plamo-3-translate-31b | 31.8B | 71.0 GB | 35.5 GB | 17.7 GB | RTX PRO 6000 | $1.38/hr | 8.0 GB |
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 (3)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| plamo-3-translate-0.8b | 890M | BF16 | 2.0 GB | RTX 3070 | $0.132/hr |
| plamo-3-translate-8b | 8.5B | BF16 | 19.1 GB | RTX 3090 | $0.162/hr |
| plamo-3-translate-31b | 31.8B | BF16 | 71.0 GB | RTX PRO 6000 | $1.38/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.
