Update perf test all-in-one config for batch_size arg (#10012)

python/llm/test/benchmark/arc-perf-test.yaml

@@ -22,6 +22,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/arc-perf-transformers-434.yaml

@@ -6,6 +6,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/core-perf-test.yaml

@@ -13,6 +13,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/cpu-perf-test.yaml

@@ -5,6 +5,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/igpu-perf/1024-128.yaml

@@ -18,6 +18,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '1024-128'
 test_api:

python/llm/test/benchmark/igpu-perf/1024-128_434.yaml

@@ -5,6 +5,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '1024-128'
 test_api:

python/llm/test/benchmark/igpu-perf/2048-256.yaml

@@ -18,6 +18,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '2048-256'
 test_api:

python/llm/test/benchmark/igpu-perf/2048-256_434.yaml

@@ -5,6 +5,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '2048-256'
 test_api:

python/llm/test/benchmark/igpu-perf/32-32.yaml

@@ -18,6 +18,7 @@ warm_up: 3
 num_trials: 5
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
 test_api:

python/llm/test/benchmark/igpu-perf/32-32_434.yaml

@@ -5,6 +5,7 @@ warm_up: 3
 num_trials: 5
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
 test_api:

python/llm/test/benchmark/igpu-perf/32-512.yaml

@@ -18,6 +18,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-512'
 test_api:

python/llm/test/benchmark/igpu-perf/32-512_434.yaml

@@ -5,6 +5,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-512'
 test_api:

python/llm/test/benchmark/stable-version-arc-perf-test-fp8.yaml

@@ -9,6 +9,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'fp8' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/stable-version-arc-perf-test-sym_int4.yaml

@@ -9,6 +9,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/stable-version-arc-stress-test-fp8.yaml

@@ -9,6 +9,7 @@ warm_up: 10
 num_trials: 100
 num_beams: 1 # default to greedy search
 low_bit: 'fp8' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '1024-512'
   - '2048-512'

python/llm/test/benchmark/stable-version-arc-stress-test-sym_int4.yaml

@@ -9,6 +9,7 @@ warm_up: 10
 num_trials: 100
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '1024-512'
   - '2048-512'

python/llm/test/benchmark/stable-version-cpu-perf-test.yaml

@@ -11,6 +11,7 @@ warm_up: 1
 num_trials: 3
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '32-32'
   - '1024-128'

python/llm/test/benchmark/stable-version-cpu-stress-test.yaml

@@ -11,6 +11,7 @@ warm_up: 3
 num_trials: 50
 num_beams: 1 # default to greedy search
 low_bit: 'sym_int4' # default to use 'sym_int4' (i.e. symmetric int4)
+batch_size: 1 # default to 1
 in_out_pairs:
   - '1024-512'
   - '2048-512'