-
Notifications
You must be signed in to change notification settings - Fork 578
Pull requests: AI-Hypercomputer/maxtext
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add foundational AbstractTrainingEngine contract methods and logging utilities
#4740
opened Aug 5, 2026 by
copybara-service
Bot
Loading…
Add train_v2.py pre-training driver and automatic NNX parameter sharding extraction
#4739
opened Aug 5, 2026 by
copybara-service
Bot
Loading…
Encapsulate lifecycle management and unify metrics logging in MaxTextTrainingEngine
#4738
opened Aug 5, 2026 by
copybara-service
Bot
Loading…
[maxtext] Add block-diffusion training primitives
#4737
opened Aug 5, 2026 by
ethannnnnn
Loading…
4 tasks done
add static LHS scaling support for GMM v2 FWD
#4735
opened Aug 4, 2026 by
CaptainO5
Collaborator
Loading…
4 tasks done
Copy the merged model in the SFT train step
#4734
opened Aug 4, 2026 by
ecnal-cienet
Collaborator
•
Draft
4 tasks done
Add indexer_cutoff_threshold field to Config and remat policy
gemini-review
#4733
opened Aug 4, 2026 by
zcjhao
Collaborator
Loading…
Fix Qwen3.5 GDN Kernel ImportError on MaxText
pull ready
#4732
opened Aug 4, 2026 by
niting
Collaborator
Loading…
4 tasks done
Fix Qwen3/3.5 KV Cache list indexing in vLLM
#4731
opened Aug 4, 2026 by
niting
Collaborator
Loading…
4 tasks done
Add filelock to tokenizer asset downloads in MaxText post-training tests.
#4730
opened Aug 4, 2026 by
copybara-service
Bot
Loading…
1 task done
Add Self-Patching Agent for the Checkpoint Validation Pipeline
#4729
opened Aug 4, 2026 by
olufiyin19
Collaborator
Loading…
4 tasks done
Add Decoding Verification for the Checkpoint Validation Pipeline
#4728
opened Aug 4, 2026 by
olufiyin19
Collaborator
Loading…
4 tasks done
Add
layer_metrics.py tool for Layer by Layer Activation for Forward Pass Debugging in Checkpoint Validation Pipeline
#4727
opened Aug 4, 2026 by
olufiyin19
Collaborator
Loading…
4 tasks done
Add Forward Pass Logits Verification for the Checkpoint Validation Pipeline
#4726
opened Aug 4, 2026 by
olufiyin19
Collaborator
Loading…
4 tasks done
test(unit): remove skipped test markers for sharding tests
pull ready
#4725
opened Aug 4, 2026 by
RexBearIU
Collaborator
Loading…
4 tasks done
Fix Qwen3-VL-2B model E2E framework stability issues when running in text-only mode (
use_multimodal=false).
#4724
opened Aug 4, 2026 by
chiajunglien
Contributor
Loading…
4 tasks done
feat(ci): implement domain-relevance gating in analyze_code_changes
#4723
opened Aug 4, 2026 by
RexBearIU
Collaborator
Loading…
4 tasks done
fix(ci): enable docker --init and add 500ms trap handlers for TPU runner lock release on cancellation
#4722
opened Aug 4, 2026 by
RexBearIU
Collaborator
Loading…
4 tasks done
fix(ci): enforce post_training isolation in test coordinator and fix missing test markers across test files
#4721
opened Aug 4, 2026 by
RexBearIU
Collaborator
Loading…
4 tasks done
Avoid re-gathering vocab tiling output head in backward
gemini-review
#4720
opened Aug 4, 2026 by
huytransformer
Collaborator
•
Draft
4 tasks
Revert PR #4636: unconditionally update rollout params before pre-RL evaluation
#4718
opened Aug 3, 2026 by
igorts-git
Collaborator
Loading…
4 tasks done
Expose the training mesh on MaxTextTrainingEngine
#4716
opened Aug 3, 2026 by
haochengd-google
Loading…
4 tasks done
[MaxText][vLLM] Fix weight converter layout transposes for Qwen3.5 MoE
#4713
opened Aug 3, 2026 by
igorts-git
Collaborator
Loading…
4 tasks done
Enable 3D Mrope for Qwen3 in multimodal training
#4709
opened Aug 3, 2026 by
lkhphuc
Contributor
Loading…
2 of 4 tasks
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.