-
[Model] Add support for OLMo architecture (#3046)
发布于
2024-12-14 01:38:19 +00:00 | 203 次提交 在此版本后已推送到 mainThis PR add support for OLMo architecture.
Additional support: add support for clip-qkv.
Test: already tested on android(pixel 4) and cuda(setting tensor_parallel_shrads=2)
下载附件