Update on "Add W4A8 INT8 activation kernels for batched MoE prefill"

INT8 tensor core variants of the batched MoE GEMM kernels that
dynamically quantize bf16 activations to INT8 per-row per-tile and
dequantize INT4 weights directly to INT8 (skipping bf16 conversion).
Uses tl.dot(int8, int8) → int32 accumulation with per-tile float32
rescale. 1.7× MoE speedup on A100 at M=1024 with 0.9998 cosine
similarity vs bf16 baseline.

Co-authored-by: Claude <noreplyanthropic.com>

[ghstack-poisoned]

Author: digantdesai

.ci/scripts/export_model_artifact.sh

@@ -418,7 +418,8 @@ if [ "$MODEL_NAME" = "qwen3_5_moe" ]; then
   TORCHINDUCTOR_CACHE_DIR="$INDUCTOR_CACHE" \
   python -m executorch.examples.models.qwen3_5_moe.export \
       --prequantized "$LOCAL_MODEL_DIR" \
-      --output-dir "${OUTPUT_DIR}"
+      --output-dir "${OUTPUT_DIR}" \
+      --moe-activation-dtype int8
   echo "::endgroup::"
 
   test -f "${OUTPUT_DIR}/model.pte"

examples/models/qwen3_5_moe/export.py

@@ -952,7 +952,7 @@ def main():  # noqa: C901
         "--moe-activation-dtype",
         choices=["bf16", "int8"],
         default="bf16",
-        help="MoE activation dtype for prefill only. Decode always uses bf16. bf16 (default): W4A16 batched GEMM. int8: W4A8 with INT8 tensor cores (~1.5x faster prefill).",
+        help="MoE activation dtype for prefill only. Decode always uses bf16. bf16 (default): W4A16 batched GEMM. int8: W4A8 with INT8 tensor cores.",
     )
     args = parser.parse_args()