Can I run GLiDE on an RTX 3090?
Probably, at 4-bit
The published weights are too big, but a 4-bit version (about 14.7 GB) would fit, with room for ≈ 101K tokens of context. The site hasn't analyzed a GGUF of GLiDE yet.
Files and context on RTX 3090
Each file of GLiDE (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 RTX 3090 | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| As published | 52.2 GB | Doesn't fit | — | — |
| 4-bit (estimate) | 14.7 GB | Fits | 101K | 202K |