sensenova models
4 sensenova models from sensenova on Hugging Face, from 9.7B to 27.4B parameters, published by sensenova in F32 and BF16, with a 256K-token context. The smallest official model, SenseNova-U1.5-8B-MoT, needs about 39.2 GB of VRAM at its published precision; the cheapest live fit is RTX A6000 at $0.363/hr.
Pick a size
One row per official sensenova 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 |
|---|---|---|---|---|---|---|---|
| SenseNova-Loop-9B-Preview | 9.7B | 43.2 GB | 10.8 GB | 5.4 GB | RTX A6000 | $0.363/hr | 1.0 GB |
| SenseNova-U1.5-8B-MoT | 17.5B | 39.2 GB | 19.6 GB | 9.8 GB | RTX A6000 | $0.363/hr | not published for this architecture |
| SenseNova-U1-8B-MoT | 17.6B | 39.2 GB | 19.6 GB | 9.8 GB | RTX A6000 | $0.363/hr | not published for this architecture |
| SenseNova-6.8-Code2Video | 27.4B | 61.1 GB | 30.6 GB | 15.3 GB | A100 | $1.31/hr | 2.0 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 (4)
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| SenseNova-Loop-9B-Preview | 9.7B | F32 | 43.2 GB | RTX A6000 | $0.363/hr |
| SenseNova-U1.5-8B-MoT | 17.5B | BF16 | 39.2 GB | RTX A6000 | $0.363/hr |
| SenseNova-U1-8B-MoT | 17.6B | BF16 | 39.2 GB | RTX A6000 | $0.363/hr |
| SenseNova-6.8-Code2Video | 27.4B | BF16 | 61.1 GB | A100 | $1.31/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.