| feat(torch): Qwen3-Omni support ulysses cp / fix(torch): repeat_kv and activation_offload bug
Co-authored-by: yaoyaoxu<xuyaoyao.824404@huawei.com>
# message auto-generated for no-merge-commit merge:
!2188 merge qwen3omni_ulysses_cp into master
feat(torch): Qwen3-Omni support ulysses cp / fix(torch): repeat_kv and activation_offload bug
Created-by: yaoyaoxu
Commit-by: yaoyaoxu
Merged-by: ascend-robot
Description: ## Motivation
当前序列长度无法支持128K,需要适配CP以支持128K长序列
## Modification
1.Qwen3-Omni支持ulysses cp:对AuT、ViT、LLM都做了适配;
如果开启CP,但没有传入音频数据或CP size > = seq_len,则不对音频模块做CP处理
2.修复repeat_kv的bug
3.修复开启activation_offload配置时的内存泄漏bug
## Self-test (Optional)
If modifications to this PR may cause/fix function/accuracy/performance DTSs/issues, a self-inspection record needs to be attached.
## BC-breaking (Optional)
If there are compatibility issues, such as dependencies on cann/torch_npu versions, they need to be explained in the PR.
## Checklist
**Before PR**:
- [ ] The new code needs to comply with the Clean Code specification.
- [ ] The PR content is self-checked, and the expression can be clear and the writing standardized
**After PR**:
- [ ] CLA has been signed and all committers have signed the CLA in this PR.
- [ ] The ci-pipeline is passed, Code Check is passed.
See merge request: Ascend/MindSpeed-MM!2188 | 2 个月前 |