Skip to content

Pull requests: tenstorrent/tt-inference-server

New pull request New
92 Open 1,735 Closed
92 Open 1,735 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

Add DEPLOYMENT.md for cloud team
#3503 opened May 12, 2026 by dmadicTT Collaborator Loading…
Add device liveness
#3502 opened May 12, 2026 by ddjukicTT Collaborator • Draft
2
Collapse same heading blocks in v2 reports
#3501 opened May 12, 2026 by ddjukicTT Collaborator Loading…
2
cpp server - tle integration
#3500 opened May 12, 2026 by knovokmetTT Contributor Loading…
4
Fix incosistent metric naming
#3490 opened May 12, 2026 by vpetrovicTT Collaborator • Draft
2
Forge SDXL: implement LoRA (load/fuse/unload) via recompile-on-change
#3488 opened May 12, 2026 by ctr-lelanchelianTT Collaborator Loading…
4 tasks
1
SDXL: perf benchmark script + methodology for Blackhole comparison
#3487 opened May 12, 2026 by ctr-lelanchelianTT Collaborator Loading…
2 tasks
4
SDXL: per-stage perf logging in trace and Forge runners
#3486 opened May 12, 2026 by ctr-lelanchelianTT Collaborator Loading…
3 tasks
2
Forge SDXL: support full-on-device (text encoders + VAE) via TTXLA_SDXL_FULL_ON_DEVICE
#3485 opened May 12, 2026 by ctr-lelanchelianTT Collaborator Loading…
3 tasks
2
SDXL: add Blackhole accuracy eval, trace non-LoRA eval, and LoRA rollback test
#3484 opened May 12, 2026 by ctr-lelanchelianTT Collaborator Loading…
3 tasks
1
correction of the links in models_by_hardware.md for N300 and P150 #3447
#3465 opened May 12, 2026 by vcankovicTT Contributor Loading…
1
4
Hapaic/upgrade vllm to v0.18.1 support prefixcaching community
#3457 opened May 12, 2026 by jeyoon321 Loading…
Remove BOS Token on Continuitation
#3452 opened May 11, 2026 by ljovanovicTT Contributor • Draft
Acvejic/run py pre and post 0.11 images
#3451 opened May 11, 2026 by acvejicTT Collaborator • Draft
2
patch curl request for structured output benchmarks
#3450 opened May 11, 2026 by lukezTT Contributor • Draft
2
fix(model_spec): add max_num_batched_tokens field to DeviceModelSpec community
#3433 opened May 10, 2026 by Dev-X25874 Loading…
1
cpp server - propagate cancel to blaze + handle all cases
#3428 opened May 9, 2026 by knovokmetTT Contributor • Draft
12
Fix Forge LLM models so local launches work without --override-docker-image
#3414 opened May 8, 2026 by kmabeeTT Contributor Loading…
2
Fix LLM p300x2 trace region size & set has_builtin_warmup=True
#3411 opened May 7, 2026 by bgoelTT Collaborator Loading…
1
@bgoelTT
2
ci: migrate cpp_server CI benchmarks to pytest
#3398 opened May 7, 2026 by dmadicTT Collaborator • Draft
3 of 7 tasks
2
Migrate image models to cpp_server (SDXL)
#3391 opened May 6, 2026 by ztorlakTT Contributor Loading…
1
@ztorlakTT
11
Add p300x2 model_spec.py entries and update models-ci-config.json
#3371 opened May 6, 2026 by bgoelTT Collaborator Loading…
1
@bgoelTT
5
Fix build_exabox.sh --blaze: cmake cache corruption and Rust version …
#3369 opened May 5, 2026 by dmadicTT Collaborator • Draft
v2 LLM module refactor
#3352 opened May 5, 2026 by ddjukicTT Collaborator Loading…
3
@ddjukicTT
14
Dmadic tt/glue testing
#3338 opened May 5, 2026 by dmadicTT Collaborator • Draft
2
ProTip! Follow long discussions with comments:>50.