| .. |
|
fused_moe
|
22a4cd4595
core: fix spec decode metrics and envs circular import (#889)
|
1 month ago |
|
mamba
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
ops
|
8a71788372
Add OLMoE (#772)
|
3 months ago |
|
__init__.py
|
07aa2a492f
upstream: add option to specify tokenizer
|
1 year ago |
|
activation.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
layernorm.py
|
bfc3da41ae
feat: add torch.compile for GemmaRMSNorm (#898)
|
1 month ago |
|
linear.py
|
afc9a28aa0
chore: add AphroditeParameter support for FP8 quant (#902)
|
1 month ago |
|
logits_processor.py
|
0e558e9b2f
fix: loading chameleon model with TP>1 (#695)
|
4 months ago |
|
pooler.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
rejection_sampler.py
|
e3a53712f2
fix: mlpspeculator with padded vocab (#669)
|
4 months ago |
|
rotary_embedding.py
|
4434c4db84
chore: refactor llama3 rope (#748)
|
4 months ago |
|
sampler.py
|
d49ead7334
reduce sampler peak memory usage
|
1 month ago |
|
spec_decode_base_sampler.py
|
09b82f9963
feat: Add support for GPU device selection in SpecDecodeBaseSampler (#629)
|
4 months ago |
|
typical_acceptance_sampler.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
vocab_parallel_embedding.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
50h100a
d49ead7334
reduce sampler peak memory usage