This should fit on 2x3090s on windows with a 18-19,24 gpu split with 6k-8k context. Uses the new exlv2 quant
- Downloads last month
- 6
Inference API (serverless) is not available, repository is disabled.
This should fit on 2x3090s on windows with a 18-19,24 gpu split with 6k-8k context. Uses the new exlv2 quant