Alexander Korolev
9cf76da709
mxfp4 probe: KV pin 4->8 GiB - engine needs 6.58 GiB/rank for one 131072-token request (51.6 KB/token/rank measured)
|
преди 2 месеца | |
|---|---|---|
| .. | ||
| recipe.toml | преди 2 месеца | |
|
Alexander Korolev
9cf76da709
mxfp4 probe: KV pin 4->8 GiB - engine needs 6.58 GiB/rank for one 131072-token request (51.6 KB/token/rank measured)
|
преди 2 месеца | |
|---|---|---|
| .. | ||
| recipe.toml | преди 2 месеца | |