 AlpinDale
|
64c05b969a
fix: `ShardedStateLoader` with fp8 quant (#900)
|
vor 1 Monat |
|
AlpinDale
|
0f1af04cf5
frontend: minor logging improvements (#787)
|
vor 2 Monaten |
 50h100a
|
1a5e612eaf
pass tensor progress bar as weight iterator
|
vor 3 Monaten |
|
50h100a
|
a061d7578d
gather estimated weight bytes alongside iterator and pass it to model.load_weights
|
vor 3 Monaten |
|
AlpinDale
|
7222b84582
feat: ministral support (#776)
|
vor 3 Monaten |
|
AlpinDale
|
89a2c6dee1
chore: refactor `MultiModalConfig` initialization and profiling (#745)
|
vor 4 Monaten |
|
AlpinDale
|
0b8b407b6d
feat: support profiling with multiple multi-modal inputs per prompt (#712)
|
vor 4 Monaten |
|
AlpinDale
|
3693028340
feat: support for Audio modality (#698)
|
vor 4 Monaten |
|
AlpinDale
|
5d37ec1016
suppress tpu import warning (#696)
|
vor 4 Monaten |
|
AlpinDale
|
3f49a55f82
feat: add INT8 W8A16 quant for TPU (#663)
|
vor 4 Monaten |
|
AlpinDale
|
30d7effc7c
feat: add siglip encoder for llava family (#626)
|
vor 4 Monaten |
|
AlpinDale
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
vor 4 Monaten |
