yangzhuxinyzxandOpenAI Codex ac26d0cc1e [Kernel] Reduce DFlash2 scale and LM-head residency ...

Co-authored-by: OpenAI Codex <noreply@openai.com>
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>

2026-09-12 23:45:29 +08:00

..

test_gdn_full_forward_device_gate.py

[Perf][SM70] Arm the Qwen GDN full-forward wrapper on Volta only

2026-09-08 20:24:34 +02:00

test_gdn_prefill_backend_resolve.py

[Bugfix][SM70] Make GDN prefill startable on Turing and fuse the spec-decode core

2026-09-08 20:24:34 +02:00

test_pooler_activations.py

…

test_pooler_methods.py

…

test_rocm_unquantized_gemm.py

…

test_sm70_lm_head.py

[Kernel] Reduce DFlash2 scale and LM-head residency

2026-09-12 23:45:29 +08:00