Skip to content

Pull requests: nod-ai/shark-ai

New pull request New
77 Open 917 Closed
77 Open 917 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 --use-attention-mask for tp8 as workaround for sharded compilation
#1188 opened Mar 28, 2025 by aviator19941 Loading…
[sharktank][kernels] restores unmasked attn functionality
#1187 opened Mar 28, 2025 by dan-garvey Loading…
1
Make move_shards_to_new_devices help function more useful
#1185 opened Mar 27, 2025 by Alex-Vasile Loading…
[tuner] expose the function merge_td_specs as utility executable
#1184 opened Mar 27, 2025 by bangtianliu Loading…
9
Rework batcher for a reservation system
#1183 opened Mar 27, 2025 by rsuderman Loading…
Check output paths before performing computation
#1182 opened Mar 27, 2025 by Alex-Vasile Loading…
Test PR
#1181 opened Mar 27, 2025 by renxida • Draft
[WIP] Expand split fix
#1180 opened Mar 27, 2025 by Alex-Vasile Loading…
3
[sharktank] Move expensive VAE tests to nightly
#1174 opened Mar 27, 2025 by sogartar • Draft
[sharktank] run unit tests job on MI300x machine as well
#1173 opened Mar 27, 2025 by sogartar • Draft
5
[sharktank] Perplexity refactor + add features
#1172 opened Mar 27, 2025 by archana-ramalingam Loading…
[tuner] add the option of providing starter td spec
#1171 opened Mar 27, 2025 by bangtianliu Loading…
10
[shortfin] Fix shortfin broken build due to iree changes
#1170 opened Mar 27, 2025 by zphoenixrises • Draft
@zphoenixrises
Disable info logging on save
#1164 opened Mar 26, 2025 by rsuderman Loading…
1
Shortfin llm beam search decode
#1161 opened Mar 26, 2025 by stbaione • Draft
Update README.md
#1159 opened Mar 26, 2025 by pdhirajkumarprasad Loading…
Relayout KV cache for internal reduction dimensions
#1158 opened Mar 26, 2025 by rsuderman Loading…
1
Fix the missing of --use-hf in llama benchmark test
#1157 opened Mar 25, 2025 by AmosLewis Loading…
1
[WIP] Fp8 Flux
#1155 opened Mar 25, 2025 by KyleHerndon Loading…
Bump IREE requirement pins to 3.4.0rc20250325 and remove dependencies on --iree-hal-target-backends
#1145 opened Mar 25, 2025 by shark-pr-automator bot Loading…
@renxida
1
docs: adds port forwarding tip to sdxl_rdna_demo.md
#1134 opened Mar 21, 2025 by monorimet Loading…
5
(shortfin-apps)(flux) Align flux response handling with shark-ui compatibility requirements
#1128 opened Mar 20, 2025 by monorimet Loading…
21
Bump IREE requirement pins to 3.3.0rc20250320
#1126 opened Mar 20, 2025 by shark-pr-automator bot Loading…
@renxida
1
Correcting usage of --iree-hal-target-device=llvm-cpu. bug Something isn't working
#1122 opened Mar 19, 2025 by benvanik Loading…
1
2
Add attn dtype to config json and sfnp float8 support
#1118 opened Mar 19, 2025 by aviator19941 Loading…
3
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.