Skip to content

Pull requests: nod-ai/shark-ai

New pull request New
129 Open 1,454 Closed
129 Open 1,454 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

Expand .create() and .add_to_archive() to handle ShardedTensors with InferenceTensor shards
#1884 opened Jul 18, 2025 by Alex-Vasile Loading…
Use int8 for handling float8_e4m3fn compatiblity
#1883 opened Jul 18, 2025 by KyleHerndon Loading…
Add an MLIR kernel for properly handling FP4 casting
#1882 opened Jul 18, 2025 by KyleHerndon Loading…
1
[sharktank][llm] Restrict page dimension to device_block_count
#1880 opened Jul 18, 2025 by Groverkss Loading…
1
[sharktank] Add sharding of fp4 quantized toy Llama theta
#1876 opened Jul 18, 2025 by sogartar Loading…
2
[sharktank] Make sharded-split tensor support quantized tensors
#1875 opened Jul 18, 2025 by sogartar Loading…
6
Bump IREE requirement pins to 3.6.0rc20250718
#1874 opened Jul 18, 2025 by shark-pr-automator bot Loading…
[sharktank] Add FP4 quantized tensor split and cat
#1873 opened Jul 18, 2025 by sogartar Loading…
4
Refactor decoder with stateful tools
#1871 opened Jul 18, 2025 by rsuderman Loading…
1
Minor fix for token selector reservation
#1867 opened Jul 17, 2025 by rsuderman Loading…
1
[sharktank] Yet another refactor of assert_tensor_close for tree support and integer dtypes
#1858 opened Jul 17, 2025 by sogartar Loading…
2
[sharktank] Add toy Llama FP4 quantization
#1857 opened Jul 17, 2025 by sogartar Loading…
2
Add native scorer
#1842 opened Jul 15, 2025 by zeeshanhaque21 • Draft
Fix view override for QuantizedTensor
#1835 opened Jul 15, 2025 by paulzzy Loading…
7
Cleanup scheduler to update rather than add / remove
#1832 opened Jul 15, 2025 by rsuderman • Draft
1
Start adding fp4 quantization accuracy test
#1829 opened Jul 15, 2025 by aviator19941 • Draft
Lisal.mooncake update write back
#1827 opened Jul 15, 2025 by lisaliu1 Loading…
1
Bump aiohttp from 3.11.3 to 3.12.14 in /shortfin dependencies Pull requests that update a dependency file python Pull requests that update python code
#1824 opened Jul 15, 2025 by dependabot bot Loading…
1
Set required Python version to ">=3.11"
#1815 opened Jul 14, 2025 by marbre Loading…
1
4
shortfin_apps.sd / builder: Change driver name from amdgpu to hip.
#1811 opened Jul 12, 2025 by monorimet Loading…
1
Sharktank extend rotary mask
#1809 opened Jul 11, 2025 by stbaione Loading…
5
Rewrite token selection strategy
#1793 opened Jul 9, 2025 by zeeshanhaque21 • Draft
Add a paramter that enables QuaRot for GEMMs
#1779 opened Jul 9, 2025 by KyleHerndon Loading…
2
[tuner] add support for attention op
#1772 opened Jul 8, 2025 by bangtianliu Loading…
16
Cast attn_output to h dtype in paged_llama_attention_block
#1761 opened Jul 4, 2025 by sebvince Loading…
1
ProTip! What’s not been updated in a month: updated:<2025-06-19.