Alexander Korolev
e0eeffe3b1
lab-dflash2: use dequantized bf16 drafter variant (arch DFlashDraftModel for lab image; MXFP8 scales unsupported by its loader)
|
1 месяц назад | |
|---|---|---|
| .. | ||
| recipe.toml | 1 месяц назад | |
|
Alexander Korolev
e0eeffe3b1
lab-dflash2: use dequantized bf16 drafter variant (arch DFlashDraftModel for lab image; MXFP8 scales unsupported by its loader)
|
1 месяц назад | |
|---|---|---|
| .. | ||
| recipe.toml | 1 месяц назад | |