diff --git a/.github/workflows/benchmark.yml b/.github/workflows/benchmark.yml
index e3b5e94488..b16cb4e0ec 100644
--- a/.github/workflows/benchmark.yml
+++ b/.github/workflows/benchmark.yml
@@ -49,9 +49,9 @@ jobs:
     - name: reset process replay
       run: python3.11 test/external/process_replay/reset.py
     - name: Run Stable Diffusion
-      run: JIT=1 python3.11 examples/stable_diffusion.py --seed 0 --noshow --timing | tee sd.txt
-    - name: Run Stable Diffusion with fp16
-      run: JIT=1 python3.11 examples/stable_diffusion.py --fp16 --seed 0 --noshow --timing | tee sd_fp16.txt
+      run: JIT=1 python3.11 examples/stable_diffusion.py --fp16 --seed 0 --noshow --timing | tee sd.txt
+    - name: Run Stable Diffusion without fp16
+      run: JIT=1 python3.11 examples/stable_diffusion.py --seed 0 --noshow --timing | tee sd_no_fp16.txt
     - name: Run SDXL
       run: JIT=1 python3.11 examples/sdxl.py --seed 0 --noshow --timing | tee sdxl.txt
     - name: Run model inference benchmark
@@ -118,7 +118,7 @@ jobs:
           matmul_half.txt
           matmul_bfloat16.txt
           sd.txt
-          sd_fp16.txt
+          sd_no_fp16.txt
           sdxl.txt
           beautiful_mnist.txt
           train_cifar.txt
@@ -178,7 +178,7 @@ jobs:
     # - name: Run Tensor Core GEMM (NV) with BEAM
     #   run: BEAM=4 NV=1 HALF=1 IGNORE_BEAM_CACHE=1 DEBUG=2 python3 extra/gemm/simple_matmul.py
     - name: Run Stable Diffusion
-      run: NV=1 python3 examples/stable_diffusion.py --seed 0 --noshow --timing | tee sd.txt
+      run: NV=1 python3 examples/stable_diffusion.py --fp16 --seed 0 --noshow --timing | tee sd.txt
     - name: Run SDXL
       run: NV=1 RUN_PROCESS_REPLAY=0 python3 examples/sdxl.py --seed 0 --noshow --timing | tee sdxl.txt
     - name: Run LLaMA
@@ -358,7 +358,7 @@ jobs:
     #- name: Fuzz Padded Tensor Core GEMM
     #  run: HSA=1 M_START=12 M_STOP=20 M_STEP=1 N_START=12 N_STOP=20 N_STEP=1 K_START=28 K_STOP=36 K_STEP=1 HALF=1 TC_OPT=2 DEBUG=2 python3 ./extra/gemm/fuzz_matmul.py
     - name: Run Stable Diffusion
-      run: AMD=1 python3 examples/stable_diffusion.py --seed 0 --noshow --timing | tee sd.txt
+      run: AMD=1 python3 examples/stable_diffusion.py --fp16 --seed 0 --noshow --timing | tee sd.txt
     - name: Run SDXL
       run: AMD=1 python3 examples/sdxl.py --seed 0 --noshow --timing | tee sdxl.txt
     - name: Run LLaMA 7B