| .. |
|
apps
|
llm: fix qwen3 moe topk renormalization (#15201)
|
2026-03-17 12:57:33 +08:00 |
|
codegen
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
engine
|
llama compute gradients explicitly + 243 GB of RAM on MP=8 (#15343)
|
2026-03-18 19:54:40 +08:00 |
|
mixin
|
pad_to to mixin [pr] (#15365)
|
2026-03-19 05:02:01 -04:00 |
|
nn
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
renderer
|
limit gl*lc (#15359)
|
2026-03-19 12:38:55 +08:00 |
|
runtime
|
no gmmu mappings with GMMU=0 (#15369)
|
2026-03-20 12:18:34 +08:00 |
|
schedule
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
uop
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
viz
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
__init__.py
|
start function and add walk rewrite (#14992)
|
2026-02-25 13:56:27 +08:00 |
|
device.py
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
dtype.py
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
|
function.py
|
more Tensor(UOp) cleanups (#15364)
|
2026-03-19 03:34:30 -04:00 |
|
gradient.py
|
add test for flat llama (#15327)
|
2026-03-18 15:16:33 +08:00 |
|
helpers.py
|
better oom msg (#15362)
|
2026-03-19 14:07:01 +08:00 |
|
py.typed
|
add a single py.typed (#6083)
|
2024-08-14 17:31:46 -07:00 |
|
tensor.py
|
dtypes.index -> dtypes.weakint (#15377)
|
2026-03-20 01:08:46 -04:00 |
