All tasks / Models
PersonaPlex 7B v1
NVIDIA's open-weights English full-duplex speech-to-speech conversation model, built on Kyutai's Moshi.
- NVIDIA
- Open weights
Where it ranks
- –Voice agentsModels · Local · no result on this task’s benchmarks yet
Facts
- Vendor
- NVIDIA
- Family
- PersonaPlex
- Released
- —
- Weights
- Open: download and run it yourself
- License
- NVIDIA Open Model License
- Parameters
- 8.37B
- GPU memory to run locally
- ~24 GB (Derived: 8.37B parameters (Hugging Face safetensors count) x 2 bytes = ~16.7 GB of bf16 weights plus runtime memory.)
- Access
- Free (open weights)
- Last verified
- 18 Oct 2026
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Artificial Analysis Big Bench Audio (speech reasoning) | 19.1% | Personaplex 7B v1Big Bench Audio speech reasoning (audio in, audio out); embedded value 0.191 as percent; no last-updated date shown. | 18 Oct 2026 | artificialanalysis.ai |
Where you can use it
No tracked tool lists this model yet; as open weights it runs in local runtimes.