Add Xiaomi MiMo-V2 #45144

casinca wants to merge 12 commits into huggingface:main from casinca:xiaomi-mimo-v2
casinca
casinca init commit with v14
87f47744
casinca Merge remote-tracking branch 'upstream/main' into xiaomi-mimo-v2
e6a3345d
casinca
casinca CI fixes: missing docstring
324cbde7
casinca CI fix: removed Mixtral unused arg
5d1d45c6
casinca CI fix: shard attn_sinks (TP)
5237ef52
casinca notes
21cedb31
casinca Merge branch 'main' into xiaomi-mimo-v2
37f3b5c0
casinca refactor: using copy of gpt-oss eager attn func for sink path
eeb7ba54
casinca Merge branch 'main' into xiaomi-mimo-v2
a35118fb
vasqu
vasqu vasqu added New model
casinca
casinca v15: polish
5f5533d8
casinca Merge branch 'main' into xiaomi-mimo-v2
bad4a43e
casinca fix: exact MiMo layer pattern when fallback
67ebdb8e
github-actions

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone