Commits
Branch selector
User selector
Datepicker
Commit History
Commits on Oct 8, 2025
Bump vllm from 0.10.1.1 to 0.11.0 in /requirements (#4)
Show description for a2153d8authored
Commits on Oct 1, 2025
Bump vite from 6.3.5 to 6.3.6 in /examples/voice_agent/client (#3)
Show description for 6c69d36authoredBump vllm from 0.8.5.post1 to 0.10.1.1 in /requirements (#2)
Show description for 470f9c4authored
Commits on Sep 30, 2025
Bump transformers from 4.51.3 to 4.53.0 in /requirements (#1)
Show description for c4fae31authored
Commits on Sep 25, 2025
feat: Add complete Qwen3-Next training configuration and documentation
Show description for 0e2bc0acommittedfeat: Add FineWeb dataset preprocessing and streaming support
Show description for 10b6e04committed
Intercom Engineeringfeat: Add Qwen3-Next hybrid attention architecture support
Show description for 6846fc2committedIntercom EngineeringUpdate changelog for `r2.3.0` (#14812)
Show description for 2800752andauthoredAdd mistral small3 24B config and recipe (#14784)
Show description for a9fd59aandauthoredAdd community label bot (#14796)
Show description for 5d586b3authored
Commits on Sep 24, 2025
Disable blank Issues (#14788)
Show description for aff66aeauthoredOneLogger Integration (#13437)
Show description for 13fe7cbauthoredfix: Use shutil.copy fallback to handle file metadata permission errors (#14639)
Show description for a35d2afauthoredMXFP8 must only use E4M3 as dtype (#14793)
Show description for 7d3df0fauthored
Commits on Sep 23, 2025
Update gpt_oss.py (#14706)
Show description for 638d299authoredRemove export-deploy, automodel, and eval tutorials (#14790)
Show description for 1fb69acauthoredci: Automodel deprecation warning (#14787)
Show description for 21a5bc4andauthoredUpdate prune-distill notebooks to Qwen3 + simplify + mmlu eval (#14785)
Show description for 431fd11authoredReplace texterrors with kaldialign library (#14775)
Show description for 991e376authored
Commits on Sep 22, 2025
Adding bf16 Sortformer train and inference (#14627)
Show description for 709da78authoredAdd transducer timestamps without alignments, timestamps to streaming (#14766)
Show description for eb5426eandauthored
Commits on Sep 20, 2025
[Flux] Add cuda_graph_scope and cache images ids for full iteration cuda graph. (#14744)
Show description for 8cfedd7andauthored[Flux] Remove redundant host & device sync. (#14711)
Show description for d9a1c0aandauthored
Commits on Sep 19, 2025
Support additional Slurm parameters (#14742)
Show description for 20ed590andauthoredFix Some Failures (#14763)
Show description for 57dc705authoredData prediction objective for flow matching speech enhancement models (#14749)
Show description for de90351authoredRandomized shard slicing for tarred data (#14558)
Show description for 7fc5144authored
Commits on Sep 18, 2025
Use lhotse dataloader for ASR models to support in-manifest channel selection for multichannel recordings (#14586)
Show description for a035e05authored
Commits on Sep 17, 2025
detach arg option for run scripts (#14722)
Show description for a961bf1authoredremove env var (#14739)
Show description for 8f93234authoredcast SE weights and activations to fp32 (#14743)
Show description for 910236fauthoredimported get_moe_layer_wise_logging_tracker from megatron core moe_utils (#14694)
Show description for 350ec2dauthored
Commits on Sep 16, 2025
feat: Compatibility modification of megatron-fsdp (#14593)
Show description for cf17ca0andauthored
Commits on Sep 15, 2025
drop speech_llm example suite (#14683)
Show description for 52bfd8aauthoredUpdate ModelCommPGs API from megatron-core (#14578)
Show description for d2067cbandauthored