نویسنده | SHA1 پیام | تاریخ |
---|---|---|
|
145e554a4d neuron: add 8bit quantization for Neuron (#994) | 1 ماه پیش |
|
0dfa6b60ec core: support logprobs with multi-step scheduling (#963) | 1 ماه پیش |
|
ba6d798784 neuron: support for context length and token bucketing (#960) | 1 ماه پیش |
|
0e558e9b2f fix: loading chameleon model with TP>1 (#695) | 5 ماه پیش |
|
f1d0b77c92 [0.6.0] Release Candidate (#481) | 5 ماه پیش |