Can I run Hy3 on an NVIDIA B300?
Probably, at 4-bit
The published weights are too big, but a 4-bit version (about 168 GB) would fit, with room for ≈ 256K tokens of context. The site hasn't analyzed a GGUF of Hy3 yet.
Files and context on NVIDIA B300
Each file of Hy3 (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.
| File | Size | On NVIDIA B300 | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| As published | 598 GB | Doesn't fit | — | — |
| 4-bit (estimate) | 168 GB | Fits | 256K | 256K |