Skip to content

Pull requests: tenstorrent/tt-metal

New pull request New
1,154 Open 23,761 Closed
1,154 Open 23,761 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

refactor: remove pad_by_zero, replace with torch2tt_tensor (fixes #43919)
#43958 opened May 9, 2026 by tenstorrent-github-bot Contributor Loading…
1
10
Re-enable sentence-Bert model for CIs
#43956 opened May 9, 2026 by arginugaTT Contributor Loading…
4
[Bug fix] Disable realtime profiling for mock devices
#43955 opened May 9, 2026 by hshahTT • Draft
Bugfix. MoE: Compute integration fixes
#43954 opened May 9, 2026 by amorrisonTT Contributor Loading…
5
cmake: enforce TSan and Tracy are mutually exclusive
#43953 opened May 8, 2026 by tenstorrent-github-bot Contributor Loading…
3
test fast dispatch on sim
#43949 opened May 8, 2026 by micahwhiteTT • Draft
Performance: Fuse add into the layernorm_pre_allgather kernel
#43948 opened May 8, 2026 by fplavecTT Contributor • Draft
1
8
Fix SDPA SFPI rounding mode
#43947 opened May 8, 2026 by jasondavies Contributor Loading…
1
5
distributed: guard x86 cache-flush intrinsics for aarch64
#43940 opened May 8, 2026 by blozano-tt Contributor Loading…
2
Bug fix: Fix Blackhole DRAM adjacent reader selection
#43939 opened May 8, 2026 by jbaumanTT Contributor Loading…
3
Support Wan VAE width padding
#43938 opened May 8, 2026 by sosborne-TT Contributor Loading…
6
[Cleanup]: Moving new device apis out of experimental namespace and into api directory
#43934 opened May 8, 2026 by abhullar-tt Contributor Loading…
2
Generalize moe_compute shape support beyond hardcoded model configs Mixture of Experts (MoE) Mixture of experts module in ttnn
#43932 opened May 8, 2026 by gajanan-choudhary Contributor • Draft
5 tasks done
1
@gajanan-choudhary
2
Fix SFPU sqrt/rsqrt signed zero handling
#43929 opened May 8, 2026 by benzaid32 Loading…
4
#40888: universal_input_support_for_sort
#43925 opened May 8, 2026 by abdulla-TT Contributor • Draft
2
Switch blackhole channel 0 DRAM endpoint usage.
#43921 opened May 8, 2026 by jbaumanTT Contributor Loading…
2
fix(program-patching): handle duplicate Buffer* in tensor_buffers; re-enable muon test
#43920 opened May 8, 2026 by rmillerTT Contributor Loading…
2 of 3 tasks
1
5
Document topk fast-path shape guidance
#43918 opened May 8, 2026 by yieldthought Contributor Loading…
@cmaryanTT
2
[Bug fix] ttnn.embedding: fix partial trailing chunk in chunked tilize path
#43912 opened May 8, 2026 by sshonTT Contributor Loading…
3
aaaa
#43905 opened May 8, 2026 by jpanasiukTT Contributor • Draft
[tt-triage] Dispatcher and Inspector data hardening
#43902 opened May 8, 2026 by onenezicTT Contributor Loading…
3
11
gpt-oss MoE: pad short prefill seqs up to seq_len_per_chip and slice back
#43901 opened May 8, 2026 by sraizada-tt Contributor Loading…
3 tasks done
3
ct/fabric cluster update
#43899 opened May 8, 2026 by ctaylorTT Contributor Loading…
31
ttnn/concat: fall back to single-buffered CB when depth=2 exceeds L1
#43898 opened May 8, 2026 by kamalrajkannan78 Contributor • Draft
2 tasks done
1
[codex] Fix quad-ring MoE expert ownership
#43897 opened May 8, 2026 by yieldthought Contributor • Draft
@yieldthought
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.