[Perf][Kernel][Quantization] Fuse ReLU2 with static FP8 activation quantization (#53793)

Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: snordmann <snordmann@nvidia.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: ElizaWszola <ewszola@redhat.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Samuel Nordmann
2026-09-14 10:26:15 -07:00
committed by GitHub
co-authored by mgoin Claude Opus 4.8 ElizaWszola OpenAI Codex mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent 435c96f9db
commit 03dc26e639
12 changed files with 459 additions and 32 deletions
+2
View File
@@ -252,6 +252,8 @@ class LinearBase(PluggableLayer):
disable_tp: If true, tensor parallelism will be disabled for this layer.
"""
requires_unquantized_input = False
def __init__(
self,
input_size: int,