Add torch 2.9 in regression tests#3311

Merged

jainapurva merged 23 commits intomainfrom

add_torch2.9_tests

Nov 17, 2025

Commits on Nov 7, 2025

Add torch2.9 in regression tests
jainapurva
committed

Commits on Nov 12, 2025

Commits on Nov 13, 2025

Move dyn_int8_act_int4_wei_cpu_layout to prototype/dtypes (#3299)
jainapurva
committed
Skip quantization when channels_out / channels_in are not multiple of 16 (#3309)

jerryzh168
authored and
jainapurva
committed
[mxfp8 moe training][BE] add docs showing equivalent convergence to bf16 at scale (#3312)

danielvegamyhre
authored and
jainapurva
committed
Move marlin_qqq_tensor to prototype/dtypes (#3307)
jainapurva
committed
Enable `PerRow(axis)` to support axes other than `-1` (#3303)

vkuzo
authored and
jainapurva
committed
Remove old TORCH_VERSION variables (#3146)

andrewor14
authored and
jainapurva
committed
Add per tensor fp8 conv2d support (#3315)

jerryzh168
authored and
jainapurva
committed
Pin pytest==8.4.2 (#3321)

huydhn
authored and
jainapurva
committed
Update common used toy linear model (#3275)

namgyu-youn
authored and
jainapurva
committed
Use conda libgcc-ng 11.2 (#3327)

atalman
authored and
jainapurva
committed
Move gemlite layout to prototype/dtypes (#3313)
jainapurva
committed
Move uintx_layout to prototype/dtypes (#3316)
jainapurva
committed
Add __str__ to FqnToConfig to make printing more readable (#3323)

jcaip
authored and
jainapurva
committed
Add support for e2e benchmark for conv2d/conv3d (#3329)

authored and
jainapurva
committed
Move floatx_tensor_core_layout to prototype/dtypes (#3317)
jainapurva
committed
Use conda libgcc-ng 11.2 for nightly tests (#3326)
jainapurva
committed
Fix tests
jainapurva
committed
Merge origin/main into add_torch2.9_tests
jainapurva
committed

Commits on Nov 14, 2025