What's Changed

• Update version.txt after 0.13.4 release by @mrwyattii in https://github.com/microsoft/DeepSpeed/pull/5196
• Fix assertion to run pipeline engine with a compiled module by @tohtana in https://github.com/microsoft/DeepSpeed/pull/5197
• Allow specifying MII branch on MII CI by @mrwyattii in https://github.com/microsoft/DeepSpeed/pull/5208
• [zero++] Synchronize at the end of secondary partitioning and simplify the logic by @ByronHsu in https://github.com/microsoft/DeepSpeed/pull/5216
• Add fp16 support of Qwen1.5 models (0.5B to 72B) to DeepSpeed-FastGen by @ZonePG in https://github.com/microsoft/DeepSpeed/pull/5219
• Rename nv-torch-latest-cpu workflow to cpu-torch-latest by @loadams in https://github.com/microsoft/DeepSpeed/pull/5226
• Fix moe cpu offload by @RezaYazdaniAminabadi in https://github.com/microsoft/DeepSpeed/pull/5220
• Use deepspeed.comm instead of torch.distributed by @jinyouzhi in https://github.com/microsoft/DeepSpeed/pull/5225
• fix fused_qkv model accuracy issue by @Yejing-Lai in https://github.com/microsoft/DeepSpeed/pull/5217

Full Changelog: https://github.com/microsoft/DeepSpeed/compare/v0.13.4...v0.13.5