Skip to content

Pull requests: google-ai-edge/LiteRT

New pull request New
2,732 Open 6,799 Closed
2,732 Open 6,799 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

Expose the remaining MlDriftDelegateOptions in InferenceCalculatorOptions
#10066 opened Sep 18, 2026 by copybara-service Bot Loading…
Adjust the data pointer and size based on the offset of the weights
#10065 opened Sep 18, 2026 by copybara-service Bot Loading…
Add SwiGLU Generator to ATS.
#10064 opened Sep 18, 2026 by copybara-service Bot Loading…
Fix GPU bool options clobbering backend defaults when unset
#10063 opened Sep 18, 2026 by copybara-service Bot Loading…
Add custom_ops_gpu skill under _agents/skills/custom_ops_gpu/SKILL.md to guide
#10062 opened Sep 18, 2026 by copybara-service Bot Loading…
Add quantized (INT8/INT4/INT2 weights) and FP16/BF16 execution to the YNNPACK delegate's MoE op.
#10061 opened Sep 18, 2026 by copybara-service Bot Loading…
[LiteRT][MediaTek] Parameterize the Neuron compiler --option-bundle value
#10060 opened Sep 18, 2026 by copybara-service Bot Loading…
Copybara import of the project:
#10059 opened Sep 18, 2026 by copybara-service Bot Loading…
Copybara import of the project:
#10058 opened Sep 18, 2026 by copybara-service Bot Loading…
Relocate TFLite operator fuzzing tests to LiteRT.
#10056 opened Sep 18, 2026 by copybara-service Bot Loading…
ml_drift: Add PowerVR Rogue workarounds and subversion parsing
#10055 opened Sep 18, 2026 by copybara-service Bot Loading…
Stop restoring GPU program cache entries that describe a different model.
#10052 opened Sep 17, 2026 by copybara-service Bot Loading…
Allow explicit prefill and decode signature selection
#10051 opened Sep 17, 2026 by copybara-service Bot Loading…
Expose delegation metrics through LiteRT C and C++ compiled model APIs.
#10050 opened Sep 17, 2026 by copybara-service Bot Loading…
Prevent template injection and set least-privilege workflow permissions
#10048 opened Sep 17, 2026 by copybara-service Bot Loading…
Pin GitHub Actions to commit SHAs and container images to digests
#10047 opened Sep 17, 2026 by copybara-service Bot Loading…
Add GPU delegate options for program cache serialization, priority, and texture weight preferences to LiteRT InferenceCalculator options and OpenCL delegate.
#10046 opened Sep 17, 2026 by copybara-service Bot Loading…
Request supported WebGPU extensions for client-provided devices.
#10045 opened Sep 17, 2026 by copybara-service Bot Loading…
[MLDrift] Read a float16 constant as float32 for a float32 CONCATENATION
#10044 opened Sep 17, 2026 by huningxin Contributor Loading…
1
Relocate TFLite operator fuzzing tests to LiteRT.
#10043 opened Sep 17, 2026 by copybara-service Bot Loading…
Add BlockwiseQuantization.block_shape to the LiteRT schema.
#10042 opened Sep 17, 2026 by copybara-service Bot Loading…
Synchronize repo
#10041 opened Sep 17, 2026 by copybara-service Bot Loading…
Use private footprint bytes for memory reporting.
#10039 opened Sep 17, 2026 by copybara-service Bot Loading…
Add RepeatKVHeads to transformer_ops.
#10038 opened Sep 17, 2026 by copybara-service Bot Loading…
Automated Code Change
#10037 opened Sep 17, 2026 by copybara-service Bot Loading…
ProTip! Updated in the last three days: updated:>2026-09-15.