Skip to content

Pull requests: AI-Hypercomputer/maxtext

New pull request New
140 Open 1,065 Closed
140 Open 1,065 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

Add prefill_multisampling() and bulk_insert() in maxengine.
#1320 opened Feb 28, 2025 by Lumosis Loading…
4 tasks done
@Lumosis
fix hf pipeline unit tests
#1319 opened Feb 28, 2025 by aireenmei • Draft
4 tasks done
fixes to conversion script and llama3.1-405b checkpointing
#1318 opened Feb 27, 2025 by vivianrwu Loading…
4 tasks done
Add Llama3.1-8B benchmark with disabled collective matmul
#1317 opened Feb 27, 2025 by bvandermoon Loading…
4 tasks done
1
Refactor: Move metrics logic to MetricLogger class pull ready
#1315 opened Feb 26, 2025 by SurbhiJainUSC Loading…
4 tasks done
@gobbleturk
10
Removing SentencePiece Dependency pin as it's no longer needed
#1314 opened Feb 26, 2025 by parambole Loading…
4 tasks done
3
Add Disruption Manager & Suspend/Resume Test Script
#1312 opened Feb 26, 2025 by SujeethJinesh • Draft
4 tasks
Update models.py to fix num_layers issue for scan_layer=False in Deep…
#1311 opened Feb 25, 2025 by deval281shah Loading…
2 tasks
4
Poc elastic training
#1310 opened Feb 25, 2025 by lukebaumann • Draft
4 tasks
3
Sujinesh/add checkpointing to benchmarks
#1306 opened Feb 24, 2025 by SujeethJinesh • Draft
4 tasks
integrate gpu pallas flash attention . Reduce prefill time for llama70b pull ready
#1305 opened Feb 24, 2025 by jwyang-google Loading…
4 tasks done
@tohaowu
3
LLama 405B Tweaks
#1298 opened Feb 22, 2025 by anfals • Draft
4
create separate emergency ckpt directories for each process
#1281 opened Feb 18, 2025 by keshavb96 Loading…
@xuefgu
Prefix Caching with HBM and latency test pull ready
#1278 opened Feb 17, 2025 by yuyanpeng-google Loading…
4 tasks done
49
Link to maxtext_xpk_runner in readme for getting started with XPK pull ready
#1275 opened Feb 14, 2025 by Obliviour Loading…
4 tasks done
fix: Pin orbax-checkpoint to v0.10.3 to resolve dependency error (#1273)
#1274 opened Feb 14, 2025 by shota-inoue-lts Loading…
@A9isha
3
[WIP]chunked prefill
#1272 opened Feb 14, 2025 by mailvijayasingh Loading…
4 tasks done
7
Added support of training with NANOO fp8 GEMM on AMD MI300/MI325 GPUs.
#1262 opened Feb 11, 2025 by wenchenvincent Loading…
4 tasks done
3
BigQuery Support
#1250 opened Feb 7, 2025 by Obliviour Loading…
4 tasks done
19
Use causal mask type for training or prefill with fixed seqlen when
#1248 opened Feb 7, 2025 by wenchenvincent • Draft
3 of 4 tasks
Enable mlperf to use mixtral-v1 tokenized dataset to avoid dropping in total_weights when splitting data across multiple host
#1246 opened Feb 7, 2025 by aireenmei Loading…
4 tasks done
Upload HLOs
#1244 opened Feb 6, 2025 by Obliviour • Draft
4 tasks
1
POC for GPU e2e feature testing
#1243 opened Feb 6, 2025 by sfvaroglu • Draft
Gpu inference
#1240 opened Feb 5, 2025 by jwyang-google Loading…
4 tasks
[wip-prototype]chunked prefill on 1k prompt len
#1237 opened Feb 4, 2025 by mailvijayasingh Loading…
4 tasks
ProTip! Updated in the last three days: updated:>2025-02-25.