mlboydaisuke
/

Qwen3.8-27B-CoreAI

Image-Text-to-Text

speculative-decoding

Model card Files Files and versions

Qwen3.8-27B-CoreAI

107 GB

Ctrl+K

Ctrl+K

1 contributor

History: 15 commits

mlboydaisuke's picture

qwen3_8_27b_verify_s9_int4lin_d4_hpost: int4 S=9 verify + hidden out for the MTP drafter (29.0 tok/s code, lossless)

aaccef9 verified 6 days ago

gpu-pipelined
qwen3_8_27b_verify_s9_int4lin_d4_hpost: int4 S=9 verify + hidden out for the MTP drafter (29.0 tok/s code, lossless) 6 days ago
.gitattributes

3.46 kB
qwen3_8_27b_verify_s9_int4lin_d4_hpost: int4 S=9 verify + hidden out for the MTP drafter (29.0 tok/s code, lossless) 6 days ago
LICENSE

11.5 kB
Qwen3.8-27B: LICENSE 8 days ago
README.md

9.52 kB
int4 + lossless MTP ⚡Spec: code 29.0 tok/s, free 22.2 (M4 Max) 6 days ago
config.json

4.31 kB
Qwen3.8-27B: config.json 8 days ago