-
Notifications
You must be signed in to change notification settings - Fork 179
Pull requests: google-ai-edge/litert-torch
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Calibrate with the model's own chat template instead of Gemma's turn markers
#1182
opened Aug 22, 2026 by
john-rocky
Loading…
Decompose grouped transposed convolutions into parallel single-group transposed convolutions.
#1179
opened Aug 20, 2026 by
copybara-service
Bot
Loading…
Pass valid_mask to cache_update signatures in calibration executor
#1178
opened Aug 20, 2026 by
john-rocky
Loading…
Apply full torch_tfl decomposition table on default lowering path (M1)
#1174
opened Aug 17, 2026 by
copybara-service
Bot
Loading…
Fold dequantized constant expressions during LiteRT runtime constant folding.
#1171
opened Aug 14, 2026 by
copybara-service
Bot
Loading…
Fix KeyError: 'valid_mask' in sampling executor for split-cache models
#1169
opened Aug 14, 2026 by
copybara-service
Bot
Loading…
Add litert_lm_system_metadata to export config to support custom model metadata.
#1167
opened Aug 11, 2026 by
copybara-service
Bot
Loading…
Add model_type override to allow for tiny_gemma model type population
#1165
opened Aug 6, 2026 by
copybara-service
Bot
Loading…
Implement generic op delegation filtering for YNNPACK.
#1164
opened Aug 6, 2026 by
copybara-service
Bot
Loading…
Always populate gemma3 model type regardless of it's a vlm or not.
#1161
opened Aug 5, 2026 by
copybara-service
Bot
Loading…
Zizmor fixes - security findings in GitHub Actions workflows
#1160
opened Aug 5, 2026 by
HarishPermal
Loading…
Add experimental tfl-optimize pattern-filtering controls to convert()
#1127
opened Jul 22, 2026 by
seven713thirteen
Loading…
Bump torch from 2.12.0 to 2.13.0
dependencies
Pull requests that update a dependency file
python
Pull requests that update python code
#1125
opened Jul 17, 2026 by
dependabot
Bot
Loading…
Add gemma4 case to build_llm_metadata model type dispatch
#1097
opened Jul 8, 2026 by
javierdejesusda
Loading…
Lower PReLU, pixel_shuffle/unshuffle, and step>1 slice to GPU-clean forms
ci:run
#1090
opened Jul 2, 2026 by
john-rocky
Loading…
Add NoPE (per-layer RoPE disable) support + SmolLM3-3B example
ci:run
#1087
opened Jun 30, 2026 by
john-rocky
Loading…
Keep attention QKV-unpack view chain within rank 4 for the GPU delegate
ci:run
#1079
opened Jun 23, 2026 by
john-rocky
Loading…
fix: lora out-dims for gqa/mha models (num_query_groups factor)
#1067
opened Jun 21, 2026 by
maceip
Loading…
Map ui8 result type back to torch.uint8 in ir_element_type_to_torch_dtype
#1064
opened Jun 19, 2026 by
john-rocky
Loading…
Fix constant aliasing in inline_consts caused by data_ptr() cache key
#1063
opened Jun 19, 2026 by
john-rocky
Loading…
Add workflow for stable release, also test python 3.14 in unit tests.
#1059
opened Jun 19, 2026 by
copybara-service
Bot
Loading…
Previous Next
ProTip!
Filter pull requests by the default branch with base:main.