Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
151 Open 1,087 Closed
151 Open 1,087 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Fix duplicate model_names for llama2-70b benchmarks
#1356 opened Mar 6, 2025 by bvandermoon Loading…
4 tasks done
support hf tokenizer in grain
#1355 opened Mar 6, 2025 by aireenmei Loading…
4 tasks done
Support DiLoCo training.
#1353 opened Mar 6, 2025 by ZacharyGarrett Loading…
3 of 4 tasks
Set computation dtype for PP weights
#1350 opened Mar 6, 2025 by gobbleturk Loading…
4 tasks done
@richjames0
Attention scaling fixes
#1349 opened Mar 6, 2025 by philip-essential Loading…
4 tasks done
Adding mixtral_8x7b.yml GPU config
#1348 opened Mar 5, 2025 by parambole Loading…
4 tasks done
Add test scripts for llama2-7b int8/bf16 models
#1347 opened Mar 5, 2025 by xy12181 Loading…
4 tasks done
4
moe subchannel config file
#1346 opened Mar 5, 2025 by mailvijayasingh Loading…
4 tasks done
1
Shauryag/elastic fixes
#1345 opened Mar 5, 2025 by shauryagup • Draft
[DRAFT] [DO NOT SUBMIT]Add an executable arg to benchmark runner and metrics file changes wi…
#1344 opened Mar 5, 2025 by shauryagup • Draft
4 tasks
Add request id to Maxengine
#1343 opened Mar 5, 2025 by liurupeng Loading…
4 tasks done
1
Fix Tokenizer issue for long running checkpointing config
#1342 opened Mar 5, 2025 by SujeethJinesh • Draft
4 tasks
Fix TP + TransformerEngine issue.
#1341 opened Mar 5, 2025 by wang2yn84 Loading…
4 tasks done
2
draft stuff
#1340 opened Mar 4, 2025 by Obliviour • Draft
4 tasks
Fix config in maxengine
#1339 opened Mar 4, 2025 by tohaowu Loading…
4 tasks done
5
Generate HLO and config for train and train_compile
#1338 opened Mar 3, 2025 by hengtaoguo • Draft
4 tasks
DeepSeek rollout checkpoint generation
#1337 opened Mar 3, 2025 by gagika Loading…
4 tasks done
@RissyRan @gobbleturk
1
[Inference PagedAttention] Integrate initial paged attention implementation into maxengine (2/N) pull ready
#1336 opened Mar 2, 2025 by wyzhang Loading…
4 tasks done
4
Add a conversion script from maxtext gemma-2 to huggingface format
#1333 opened Mar 1, 2025 by hxssgaa Loading…
4 tasks done
1
Add seq parallelism for attention and MoE MLP
#1328 opened Mar 1, 2025 by suexu1025 Loading…
8
Supporting LoRA in MaxText for inference only, to support multi-LoRA adapters inferencing via JetStream.
#1327 opened Feb 28, 2025 by aman2930 Loading…
4 tasks done
48
Add pathways configs to both xpk and on-device workload configs pull ready
#1326 opened Feb 28, 2025 by dipannita08 Loading…
4 tasks done
4
Add tokenizer_type tiktoken for llama3.1 pull ready
#1325 opened Feb 28, 2025 by A9isha Loading…
4 tasks done
@khatwanimohit
1
Add Disruption Manager & Suspend/Resume Test Script
#1312 opened Feb 26, 2025 by SujeethJinesh • Draft
4 tasks
Poc elastic training
#1310 opened Feb 25, 2025 by lukebaumann • Draft
4 tasks
3
ProTip! Type g i on any issue or pull request to go back to the issue listing page.