hf-transformers-bot commited on
Commit
a8c9fcb
·
verified ·
1 Parent(s): 14f22bc

Upload benchmark results for run 18261440921

Browse files
2025-10-05/18261440921/benchmark_results/Llama-2-7b-hf/Llama-2-7b-hf_benchmark_20251005_164557.json ADDED
@@ -0,0 +1,1121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_name": "Llama-2-7b-hf",
3
+ "benchmark_scenarios": [
4
+ {
5
+ "scenario_name": "eager_eager_attn",
6
+ "metadata": {
7
+ "timestamp": "2025-10-05T16:42:37.880117",
8
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
9
+ "hardware_info": {
10
+ "gpu_name": "unknown",
11
+ "gpu_memory_total_mb": 0,
12
+ "cpu_count": 160,
13
+ "memory_total_mb": 1285717,
14
+ "python_version": "3.12.10",
15
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
16
+ "cuda_version": null
17
+ },
18
+ "config": {
19
+ "name": "eager",
20
+ "model_id": "meta-llama/Llama-2-7b-hf",
21
+ "variant": "eager",
22
+ "warmup_iterations": 3,
23
+ "measurement_iterations": 5,
24
+ "num_tokens_to_generate": 100,
25
+ "device": "cuda",
26
+ "torch_dtype": "float16",
27
+ "compile_mode": null,
28
+ "compile_options": {},
29
+ "use_cache": true,
30
+ "batch_size": 1,
31
+ "sequence_length": null,
32
+ "attn_implementation": "eager",
33
+ "sdpa_backend": null,
34
+ "custom_params": {}
35
+ }
36
+ },
37
+ "measurements": {
38
+ "latency_seconds": {
39
+ "name": "latency_seconds",
40
+ "measurements": [
41
+ 1.4258780517578125,
42
+ 1.4432681884765626,
43
+ 1.4278271484375,
44
+ 1.438541259765625,
45
+ 1.4315069580078126
46
+ ],
47
+ "mean": 1.4334043212890626,
48
+ "median": 1.4315069580078126,
49
+ "std": 0.006557950751551789,
50
+ "min": 1.4258780517578125,
51
+ "max": 1.4432681884765626,
52
+ "p25": 1.4278271484375,
53
+ "p75": 1.438541259765625,
54
+ "p90": 1.4413774169921876,
55
+ "p95": 1.442322802734375,
56
+ "p99": 1.4430791113281252,
57
+ "unit": "seconds"
58
+ },
59
+ "time_to_first_token_seconds": {
60
+ "name": "time_to_first_token_seconds",
61
+ "measurements": [
62
+ 0.015104710578918457,
63
+ 0.01526698398590088,
64
+ 0.015416780471801757,
65
+ 0.015431859970092773,
66
+ 0.01534614372253418
67
+ ],
68
+ "mean": 0.015313295745849609,
69
+ "median": 0.01534614372253418,
70
+ "std": 0.00011958102217433148,
71
+ "min": 0.015104710578918457,
72
+ "max": 0.015431859970092773,
73
+ "p25": 0.01526698398590088,
74
+ "p75": 0.015416780471801757,
75
+ "p90": 0.015425828170776366,
76
+ "p95": 0.01542884407043457,
77
+ "p99": 0.015431256790161132,
78
+ "unit": "seconds"
79
+ },
80
+ "tokens_per_second": {
81
+ "name": "tokens_per_second",
82
+ "measurements": [
83
+ 70.13222475563089,
84
+ 69.28719194286039,
85
+ 70.03648873705197,
86
+ 69.5148639784531,
87
+ 69.85645402601965
88
+ ],
89
+ "mean": 69.7654446880032,
90
+ "median": 69.85645402601965,
91
+ "std": 0.31868215336763883,
92
+ "min": 69.28719194286039,
93
+ "max": 70.13222475563089,
94
+ "p25": 69.5148639784531,
95
+ "p75": 70.03648873705197,
96
+ "p90": 70.09393034819932,
97
+ "p95": 70.1130775519151,
98
+ "p99": 70.12839531488773,
99
+ "unit": "tokens/sec"
100
+ },
101
+ "time_per_output_token_seconds": {
102
+ "name": "time_per_output_token_seconds",
103
+ "measurements": [
104
+ 0.014258780517578125,
105
+ 0.014432681884765626,
106
+ 0.014278271484375,
107
+ 0.014385412597656251,
108
+ 0.014315069580078126
109
+ ],
110
+ "mean": 0.014334043212890626,
111
+ "median": 0.014315069580078126,
112
+ "std": 6.5579507515518e-05,
113
+ "min": 0.014258780517578125,
114
+ "max": 0.014432681884765626,
115
+ "p25": 0.014278271484375,
116
+ "p75": 0.014385412597656251,
117
+ "p90": 0.014413774169921876,
118
+ "p95": 0.014423228027343752,
119
+ "p99": 0.01443079111328125,
120
+ "unit": "seconds/token"
121
+ }
122
+ },
123
+ "gpu_metrics": {
124
+ "gpu_monitoring_status": "disabled",
125
+ "gpu_monitoring_reason": "no_gpus_available"
126
+ }
127
+ },
128
+ {
129
+ "scenario_name": "compiled_compile_max-autotune_eager_attn",
130
+ "metadata": {
131
+ "timestamp": "2025-10-05T16:42:54.318196",
132
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
133
+ "hardware_info": {
134
+ "gpu_name": "unknown",
135
+ "gpu_memory_total_mb": 0,
136
+ "cpu_count": 160,
137
+ "memory_total_mb": 1285717,
138
+ "python_version": "3.12.10",
139
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
140
+ "cuda_version": null
141
+ },
142
+ "config": {
143
+ "name": "compiled",
144
+ "model_id": "meta-llama/Llama-2-7b-hf",
145
+ "variant": "compiled",
146
+ "warmup_iterations": 3,
147
+ "measurement_iterations": 5,
148
+ "num_tokens_to_generate": 100,
149
+ "device": "cuda",
150
+ "torch_dtype": "float16",
151
+ "compile_mode": "max-autotune",
152
+ "compile_options": {},
153
+ "use_cache": true,
154
+ "batch_size": 1,
155
+ "sequence_length": null,
156
+ "attn_implementation": "eager",
157
+ "sdpa_backend": null,
158
+ "custom_params": {}
159
+ }
160
+ },
161
+ "measurements": {
162
+ "latency_seconds": {
163
+ "name": "latency_seconds",
164
+ "measurements": [
165
+ 2.984635498046875,
166
+ 3.037484619140625,
167
+ 3.027998291015625,
168
+ 3.0424931640625,
169
+ 3.063652099609375
170
+ ],
171
+ "mean": 3.0312527343749998,
172
+ "median": 3.037484619140625,
173
+ "std": 0.026071672901191938,
174
+ "min": 2.984635498046875,
175
+ "max": 3.063652099609375,
176
+ "p25": 3.027998291015625,
177
+ "p75": 3.0424931640625,
178
+ "p90": 3.055188525390625,
179
+ "p95": 3.0594203125,
180
+ "p99": 3.0628057421875,
181
+ "unit": "seconds"
182
+ },
183
+ "time_to_first_token_seconds": {
184
+ "name": "time_to_first_token_seconds",
185
+ "measurements": [
186
+ 0.01646251106262207,
187
+ 0.01654026985168457,
188
+ 0.016676025390625,
189
+ 0.016277357101440428,
190
+ 0.016404073715209962
191
+ ],
192
+ "mean": 0.01647204742431641,
193
+ "median": 0.01646251106262207,
194
+ "std": 0.00013332441597921972,
195
+ "min": 0.016277357101440428,
196
+ "max": 0.016676025390625,
197
+ "p25": 0.016404073715209962,
198
+ "p75": 0.01654026985168457,
199
+ "p90": 0.016621723175048828,
200
+ "p95": 0.016648874282836915,
201
+ "p99": 0.016670595169067385,
202
+ "unit": "seconds"
203
+ },
204
+ "tokens_per_second": {
205
+ "name": "tokens_per_second",
206
+ "measurements": [
207
+ 33.50492884824271,
208
+ 32.92197740520323,
209
+ 33.025117714468344,
210
+ 32.86778132525848,
211
+ 32.640781899730165
212
+ ],
213
+ "mean": 32.992117438580586,
214
+ "median": 32.92197740520323,
215
+ "std": 0.28561523569511027,
216
+ "min": 32.640781899730165,
217
+ "max": 33.50492884824271,
218
+ "p25": 32.86778132525848,
219
+ "p75": 33.025117714468344,
220
+ "p90": 33.313004394732964,
221
+ "p95": 33.40896662148784,
222
+ "p99": 33.48573640289174,
223
+ "unit": "tokens/sec"
224
+ },
225
+ "time_per_output_token_seconds": {
226
+ "name": "time_per_output_token_seconds",
227
+ "measurements": [
228
+ 0.029846354980468746,
229
+ 0.03037484619140625,
230
+ 0.030279982910156253,
231
+ 0.030424931640625,
232
+ 0.03063652099609375
233
+ ],
234
+ "mean": 0.03031252734375,
235
+ "median": 0.03037484619140625,
236
+ "std": 0.00026071672901192025,
237
+ "min": 0.029846354980468746,
238
+ "max": 0.03063652099609375,
239
+ "p25": 0.030279982910156253,
240
+ "p75": 0.030424931640625,
241
+ "p90": 0.03055188525390625,
242
+ "p95": 0.030594203125,
243
+ "p99": 0.030628057421875,
244
+ "unit": "seconds/token"
245
+ }
246
+ },
247
+ "gpu_metrics": {
248
+ "gpu_monitoring_status": "disabled",
249
+ "gpu_monitoring_reason": "no_gpus_available"
250
+ }
251
+ },
252
+ {
253
+ "scenario_name": "eager_sdpa_default",
254
+ "metadata": {
255
+ "timestamp": "2025-10-05T16:43:35.061977",
256
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
257
+ "hardware_info": {
258
+ "gpu_name": "unknown",
259
+ "gpu_memory_total_mb": 0,
260
+ "cpu_count": 160,
261
+ "memory_total_mb": 1285717,
262
+ "python_version": "3.12.10",
263
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
264
+ "cuda_version": null
265
+ },
266
+ "config": {
267
+ "name": "eager",
268
+ "model_id": "meta-llama/Llama-2-7b-hf",
269
+ "variant": "eager",
270
+ "warmup_iterations": 3,
271
+ "measurement_iterations": 5,
272
+ "num_tokens_to_generate": 100,
273
+ "device": "cuda",
274
+ "torch_dtype": "float16",
275
+ "compile_mode": null,
276
+ "compile_options": {},
277
+ "use_cache": true,
278
+ "batch_size": 1,
279
+ "sequence_length": null,
280
+ "attn_implementation": "sdpa",
281
+ "sdpa_backend": null,
282
+ "custom_params": {}
283
+ }
284
+ },
285
+ "measurements": {
286
+ "latency_seconds": {
287
+ "name": "latency_seconds",
288
+ "measurements": [
289
+ 1.419973876953125,
290
+ 1.4269293212890626,
291
+ 1.434625732421875,
292
+ 1.421938232421875,
293
+ 1.4280462646484375
294
+ ],
295
+ "mean": 1.426302685546875,
296
+ "median": 1.4269293212890626,
297
+ "std": 0.005134357194759363,
298
+ "min": 1.419973876953125,
299
+ "max": 1.434625732421875,
300
+ "p25": 1.421938232421875,
301
+ "p75": 1.4280462646484375,
302
+ "p90": 1.4319939453125,
303
+ "p95": 1.4333098388671874,
304
+ "p99": 1.4343625537109375,
305
+ "unit": "seconds"
306
+ },
307
+ "time_to_first_token_seconds": {
308
+ "name": "time_to_first_token_seconds",
309
+ "measurements": [
310
+ 0.015214149475097655,
311
+ 0.015399380683898926,
312
+ 0.01564625358581543,
313
+ 0.015450222969055176,
314
+ 0.015266743659973145
315
+ ],
316
+ "mean": 0.015395350074768065,
317
+ "median": 0.015399380683898926,
318
+ "std": 0.00015189017047895128,
319
+ "min": 0.015214149475097655,
320
+ "max": 0.01564625358581543,
321
+ "p25": 0.015266743659973145,
322
+ "p75": 0.015450222969055176,
323
+ "p90": 0.015567841339111328,
324
+ "p95": 0.015607047462463379,
325
+ "p99": 0.01563841236114502,
326
+ "unit": "seconds"
327
+ },
328
+ "tokens_per_second": {
329
+ "name": "tokens_per_second",
330
+ "measurements": [
331
+ 70.42383076410717,
332
+ 70.08055585378384,
333
+ 69.70459105817389,
334
+ 70.3265428271648,
335
+ 70.02574249554752
336
+ ],
337
+ "mean": 70.11225259975545,
338
+ "median": 70.08055585378384,
339
+ "std": 0.25206263839920684,
340
+ "min": 69.70459105817389,
341
+ "max": 70.42383076410717,
342
+ "p25": 70.02574249554752,
343
+ "p75": 70.3265428271648,
344
+ "p90": 70.38491558933022,
345
+ "p95": 70.40437317671869,
346
+ "p99": 70.41993924662947,
347
+ "unit": "tokens/sec"
348
+ },
349
+ "time_per_output_token_seconds": {
350
+ "name": "time_per_output_token_seconds",
351
+ "measurements": [
352
+ 0.01419973876953125,
353
+ 0.014269293212890626,
354
+ 0.01434625732421875,
355
+ 0.01421938232421875,
356
+ 0.014280462646484375
357
+ ],
358
+ "mean": 0.01426302685546875,
359
+ "median": 0.014269293212890626,
360
+ "std": 5.134357194759382e-05,
361
+ "min": 0.01419973876953125,
362
+ "max": 0.01434625732421875,
363
+ "p25": 0.01421938232421875,
364
+ "p75": 0.014280462646484375,
365
+ "p90": 0.014319939453125,
366
+ "p95": 0.014333098388671875,
367
+ "p99": 0.014343625537109375,
368
+ "unit": "seconds/token"
369
+ }
370
+ },
371
+ "gpu_metrics": {
372
+ "gpu_monitoring_status": "disabled",
373
+ "gpu_monitoring_reason": "no_gpus_available"
374
+ }
375
+ },
376
+ {
377
+ "scenario_name": "eager_sdpa_math",
378
+ "metadata": {
379
+ "timestamp": "2025-10-05T16:43:53.926823",
380
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
381
+ "hardware_info": {
382
+ "gpu_name": "unknown",
383
+ "gpu_memory_total_mb": 0,
384
+ "cpu_count": 160,
385
+ "memory_total_mb": 1285717,
386
+ "python_version": "3.12.10",
387
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
388
+ "cuda_version": null
389
+ },
390
+ "config": {
391
+ "name": "eager",
392
+ "model_id": "meta-llama/Llama-2-7b-hf",
393
+ "variant": "eager",
394
+ "warmup_iterations": 3,
395
+ "measurement_iterations": 5,
396
+ "num_tokens_to_generate": 100,
397
+ "device": "cuda",
398
+ "torch_dtype": "float16",
399
+ "compile_mode": null,
400
+ "compile_options": {},
401
+ "use_cache": true,
402
+ "batch_size": 1,
403
+ "sequence_length": null,
404
+ "attn_implementation": "sdpa",
405
+ "sdpa_backend": "math",
406
+ "custom_params": {}
407
+ }
408
+ },
409
+ "measurements": {
410
+ "latency_seconds": {
411
+ "name": "latency_seconds",
412
+ "measurements": [
413
+ 1.7282635498046874,
414
+ 1.7317625732421875,
415
+ 1.7452796630859375,
416
+ 1.7467064208984375,
417
+ 1.752925048828125
418
+ ],
419
+ "mean": 1.7409874511718748,
420
+ "median": 1.7452796630859375,
421
+ "std": 0.009387565972824008,
422
+ "min": 1.7282635498046874,
423
+ "max": 1.752925048828125,
424
+ "p25": 1.7317625732421875,
425
+ "p75": 1.7467064208984375,
426
+ "p90": 1.75043759765625,
427
+ "p95": 1.7516813232421875,
428
+ "p99": 1.7526763037109374,
429
+ "unit": "seconds"
430
+ },
431
+ "time_to_first_token_seconds": {
432
+ "name": "time_to_first_token_seconds",
433
+ "measurements": [
434
+ 0.0192354793548584,
435
+ 0.019311517715454102,
436
+ 0.019334272384643556,
437
+ 0.019920856475830077,
438
+ 0.019653024673461913
439
+ ],
440
+ "mean": 0.01949103012084961,
441
+ "median": 0.019334272384643556,
442
+ "std": 0.0002581072707468262,
443
+ "min": 0.0192354793548584,
444
+ "max": 0.019920856475830077,
445
+ "p25": 0.019311517715454102,
446
+ "p75": 0.019653024673461913,
447
+ "p90": 0.01981372375488281,
448
+ "p95": 0.019867290115356446,
449
+ "p99": 0.01991014320373535,
450
+ "unit": "seconds"
451
+ },
452
+ "tokens_per_second": {
453
+ "name": "tokens_per_second",
454
+ "measurements": [
455
+ 57.861545486682914,
456
+ 57.74463632897497,
457
+ 57.29740746716992,
458
+ 57.25060536994185,
459
+ 57.047504721809155
460
+ ],
461
+ "mean": 57.440339874915765,
462
+ "median": 57.29740746716992,
463
+ "std": 0.31008347663658536,
464
+ "min": 57.047504721809155,
465
+ "max": 57.861545486682914,
466
+ "p25": 57.25060536994185,
467
+ "p75": 57.74463632897497,
468
+ "p90": 57.81478182359974,
469
+ "p95": 57.83816365514132,
470
+ "p99": 57.8568691203746,
471
+ "unit": "tokens/sec"
472
+ },
473
+ "time_per_output_token_seconds": {
474
+ "name": "time_per_output_token_seconds",
475
+ "measurements": [
476
+ 0.017282635498046876,
477
+ 0.017317625732421876,
478
+ 0.017452796630859375,
479
+ 0.017467064208984376,
480
+ 0.01752925048828125
481
+ ],
482
+ "mean": 0.01740987451171875,
483
+ "median": 0.017452796630859375,
484
+ "std": 9.387565972823929e-05,
485
+ "min": 0.017282635498046876,
486
+ "max": 0.01752925048828125,
487
+ "p25": 0.017317625732421876,
488
+ "p75": 0.017467064208984376,
489
+ "p90": 0.0175043759765625,
490
+ "p95": 0.017516813232421873,
491
+ "p99": 0.017526763037109372,
492
+ "unit": "seconds/token"
493
+ }
494
+ },
495
+ "gpu_metrics": {
496
+ "gpu_monitoring_status": "disabled",
497
+ "gpu_monitoring_reason": "no_gpus_available"
498
+ }
499
+ },
500
+ {
501
+ "scenario_name": "eager_sdpa_flash_attention",
502
+ "metadata": {
503
+ "timestamp": "2025-10-05T16:44:15.094504",
504
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
505
+ "hardware_info": {
506
+ "gpu_name": "unknown",
507
+ "gpu_memory_total_mb": 0,
508
+ "cpu_count": 160,
509
+ "memory_total_mb": 1285717,
510
+ "python_version": "3.12.10",
511
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
512
+ "cuda_version": null
513
+ },
514
+ "config": {
515
+ "name": "eager",
516
+ "model_id": "meta-llama/Llama-2-7b-hf",
517
+ "variant": "eager",
518
+ "warmup_iterations": 3,
519
+ "measurement_iterations": 5,
520
+ "num_tokens_to_generate": 100,
521
+ "device": "cuda",
522
+ "torch_dtype": "float16",
523
+ "compile_mode": null,
524
+ "compile_options": {},
525
+ "use_cache": true,
526
+ "batch_size": 1,
527
+ "sequence_length": null,
528
+ "attn_implementation": "sdpa",
529
+ "sdpa_backend": "flash_attention",
530
+ "custom_params": {}
531
+ }
532
+ },
533
+ "measurements": {
534
+ "latency_seconds": {
535
+ "name": "latency_seconds",
536
+ "measurements": [
537
+ 1.0929300537109374,
538
+ 1.0912900390625,
539
+ 1.0941787109375,
540
+ 1.2363759765625,
541
+ 1.4184599609375
542
+ ],
543
+ "mean": 1.1866469482421875,
544
+ "median": 1.0941787109375,
545
+ "std": 0.12855848511216997,
546
+ "min": 1.0912900390625,
547
+ "max": 1.4184599609375,
548
+ "p25": 1.0929300537109374,
549
+ "p75": 1.2363759765625,
550
+ "p90": 1.3456263671875,
551
+ "p95": 1.3820431640625,
552
+ "p99": 1.4111766015625,
553
+ "unit": "seconds"
554
+ },
555
+ "time_to_first_token_seconds": {
556
+ "name": "time_to_first_token_seconds",
557
+ "measurements": [
558
+ 0.01234078598022461,
559
+ 0.012174230575561523,
560
+ 0.012040634155273438,
561
+ 0.012160832405090333,
562
+ 0.015881291389465334
563
+ ],
564
+ "mean": 0.012919554901123047,
565
+ "median": 0.012174230575561523,
566
+ "std": 0.0014839492136808946,
567
+ "min": 0.012040634155273438,
568
+ "max": 0.015881291389465334,
569
+ "p25": 0.012160832405090333,
570
+ "p75": 0.01234078598022461,
571
+ "p90": 0.014465089225769045,
572
+ "p95": 0.015173190307617188,
573
+ "p99": 0.015739671173095705,
574
+ "unit": "seconds"
575
+ },
576
+ "tokens_per_second": {
577
+ "name": "tokens_per_second",
578
+ "measurements": [
579
+ 91.49716366610997,
580
+ 91.6346676140355,
581
+ 91.392748735094,
582
+ 80.88154565897528,
583
+ 70.49899380586477
584
+ ],
585
+ "mean": 85.1810238960159,
586
+ "median": 91.392748735094,
587
+ "std": 8.416367209073124,
588
+ "min": 70.49899380586477,
589
+ "max": 91.6346676140355,
590
+ "p25": 80.88154565897528,
591
+ "p75": 91.49716366610997,
592
+ "p90": 91.57966603486528,
593
+ "p95": 91.60716682445039,
594
+ "p99": 91.62916745611847,
595
+ "unit": "tokens/sec"
596
+ },
597
+ "time_per_output_token_seconds": {
598
+ "name": "time_per_output_token_seconds",
599
+ "measurements": [
600
+ 0.010929300537109374,
601
+ 0.010912900390625,
602
+ 0.010941787109375,
603
+ 0.012363759765625,
604
+ 0.014184599609375
605
+ ],
606
+ "mean": 0.011866469482421874,
607
+ "median": 0.010941787109375,
608
+ "std": 0.0012855848511216998,
609
+ "min": 0.010912900390625,
610
+ "max": 0.014184599609375,
611
+ "p25": 0.010929300537109374,
612
+ "p75": 0.012363759765625,
613
+ "p90": 0.013456263671875,
614
+ "p95": 0.013820431640624999,
615
+ "p99": 0.014111766015624999,
616
+ "unit": "seconds/token"
617
+ }
618
+ },
619
+ "gpu_metrics": {
620
+ "gpu_monitoring_status": "disabled",
621
+ "gpu_monitoring_reason": "no_gpus_available"
622
+ }
623
+ },
624
+ {
625
+ "scenario_name": "eager_sdpa_efficient_attention",
626
+ "metadata": {
627
+ "timestamp": "2025-10-05T16:44:32.524329",
628
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
629
+ "hardware_info": {
630
+ "gpu_name": "unknown",
631
+ "gpu_memory_total_mb": 0,
632
+ "cpu_count": 160,
633
+ "memory_total_mb": 1285717,
634
+ "python_version": "3.12.10",
635
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
636
+ "cuda_version": null
637
+ },
638
+ "config": {
639
+ "name": "eager",
640
+ "model_id": "meta-llama/Llama-2-7b-hf",
641
+ "variant": "eager",
642
+ "warmup_iterations": 3,
643
+ "measurement_iterations": 5,
644
+ "num_tokens_to_generate": 100,
645
+ "device": "cuda",
646
+ "torch_dtype": "float16",
647
+ "compile_mode": null,
648
+ "compile_options": {},
649
+ "use_cache": true,
650
+ "batch_size": 1,
651
+ "sequence_length": null,
652
+ "attn_implementation": "sdpa",
653
+ "sdpa_backend": "efficient_attention",
654
+ "custom_params": {}
655
+ }
656
+ },
657
+ "measurements": {
658
+ "latency_seconds": {
659
+ "name": "latency_seconds",
660
+ "measurements": [
661
+ 1.3741121826171876,
662
+ 1.3708917236328124,
663
+ 1.3793587646484375,
664
+ 1.3785235595703125,
665
+ 1.3715455322265624
666
+ ],
667
+ "mean": 1.3748863525390624,
668
+ "median": 1.3741121826171876,
669
+ "std": 0.003491395840956925,
670
+ "min": 1.3708917236328124,
671
+ "max": 1.3793587646484375,
672
+ "p25": 1.3715455322265624,
673
+ "p75": 1.3785235595703125,
674
+ "p90": 1.3790246826171875,
675
+ "p95": 1.3791917236328126,
676
+ "p99": 1.3793253564453125,
677
+ "unit": "seconds"
678
+ },
679
+ "time_to_first_token_seconds": {
680
+ "name": "time_to_first_token_seconds",
681
+ "measurements": [
682
+ 0.014844355583190917,
683
+ 0.014871515274047852,
684
+ 0.014947993278503418,
685
+ 0.014908873558044433,
686
+ 0.014737278938293456
687
+ ],
688
+ "mean": 0.014862003326416018,
689
+ "median": 0.014871515274047852,
690
+ "std": 7.148312171993032e-05,
691
+ "min": 0.014737278938293456,
692
+ "max": 0.014947993278503418,
693
+ "p25": 0.014844355583190917,
694
+ "p75": 0.014908873558044433,
695
+ "p90": 0.014932345390319825,
696
+ "p95": 0.014940169334411621,
697
+ "p99": 0.01494642848968506,
698
+ "unit": "seconds"
699
+ },
700
+ "tokens_per_second": {
701
+ "name": "tokens_per_second",
702
+ "measurements": [
703
+ 72.77426200351131,
704
+ 72.94522118421118,
705
+ 72.49745502250633,
706
+ 72.54137900346812,
707
+ 72.91044857815281
708
+ ],
709
+ "mean": 72.73375315836995,
710
+ "median": 72.77426200351131,
711
+ "std": 0.18462155255924387,
712
+ "min": 72.49745502250633,
713
+ "max": 72.94522118421118,
714
+ "p25": 72.54137900346812,
715
+ "p75": 72.91044857815281,
716
+ "p90": 72.93131214178783,
717
+ "p95": 72.93826666299951,
718
+ "p99": 72.94383027996885,
719
+ "unit": "tokens/sec"
720
+ },
721
+ "time_per_output_token_seconds": {
722
+ "name": "time_per_output_token_seconds",
723
+ "measurements": [
724
+ 0.013741121826171876,
725
+ 0.013708917236328125,
726
+ 0.013793587646484376,
727
+ 0.013785235595703126,
728
+ 0.013715455322265624
729
+ ],
730
+ "mean": 0.013748863525390625,
731
+ "median": 0.013741121826171876,
732
+ "std": 3.4913958409569503e-05,
733
+ "min": 0.013708917236328125,
734
+ "max": 0.013793587646484376,
735
+ "p25": 0.013715455322265624,
736
+ "p75": 0.013785235595703126,
737
+ "p90": 0.013790246826171876,
738
+ "p95": 0.013791917236328126,
739
+ "p99": 0.013793253564453127,
740
+ "unit": "seconds/token"
741
+ }
742
+ },
743
+ "gpu_metrics": {
744
+ "gpu_monitoring_status": "disabled",
745
+ "gpu_monitoring_reason": "no_gpus_available"
746
+ }
747
+ },
748
+ {
749
+ "scenario_name": "compiled_compile_max-autotune_sdpa_default",
750
+ "metadata": {
751
+ "timestamp": "2025-10-05T16:44:49.777043",
752
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
753
+ "hardware_info": {
754
+ "gpu_name": "unknown",
755
+ "gpu_memory_total_mb": 0,
756
+ "cpu_count": 160,
757
+ "memory_total_mb": 1285717,
758
+ "python_version": "3.12.10",
759
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
760
+ "cuda_version": null
761
+ },
762
+ "config": {
763
+ "name": "compiled",
764
+ "model_id": "meta-llama/Llama-2-7b-hf",
765
+ "variant": "compiled",
766
+ "warmup_iterations": 3,
767
+ "measurement_iterations": 5,
768
+ "num_tokens_to_generate": 100,
769
+ "device": "cuda",
770
+ "torch_dtype": "float16",
771
+ "compile_mode": "max-autotune",
772
+ "compile_options": {},
773
+ "use_cache": true,
774
+ "batch_size": 1,
775
+ "sequence_length": null,
776
+ "attn_implementation": "sdpa",
777
+ "sdpa_backend": null,
778
+ "custom_params": {}
779
+ }
780
+ },
781
+ "measurements": {
782
+ "latency_seconds": {
783
+ "name": "latency_seconds",
784
+ "measurements": [
785
+ 1.691375,
786
+ 1.6944483642578125,
787
+ 1.6897535400390624,
788
+ 1.6890848388671875,
789
+ 1.6840247802734376
790
+ ],
791
+ "mean": 1.6897373046875,
792
+ "median": 1.6897535400390624,
793
+ "std": 0.003403969407212507,
794
+ "min": 1.6840247802734376,
795
+ "max": 1.6944483642578125,
796
+ "p25": 1.6890848388671875,
797
+ "p75": 1.691375,
798
+ "p90": 1.6932190185546876,
799
+ "p95": 1.69383369140625,
800
+ "p99": 1.6943254296875,
801
+ "unit": "seconds"
802
+ },
803
+ "time_to_first_token_seconds": {
804
+ "name": "time_to_first_token_seconds",
805
+ "measurements": [
806
+ 0.017334369659423827,
807
+ 0.017626760482788086,
808
+ 0.017680519104003905,
809
+ 0.017632482528686523,
810
+ 0.017587324142456055
811
+ ],
812
+ "mean": 0.017572291183471683,
813
+ "median": 0.017626760482788086,
814
+ "std": 0.0001225853391320522,
815
+ "min": 0.017334369659423827,
816
+ "max": 0.017680519104003905,
817
+ "p25": 0.017587324142456055,
818
+ "p75": 0.017632482528686523,
819
+ "p90": 0.01766130447387695,
820
+ "p95": 0.017670911788940428,
821
+ "p99": 0.01767859764099121,
822
+ "unit": "seconds"
823
+ },
824
+ "tokens_per_second": {
825
+ "name": "tokens_per_second",
826
+ "measurements": [
827
+ 59.12349419850713,
828
+ 59.016256918399,
829
+ 59.18022814006845,
830
+ 59.20365732905793,
831
+ 59.381548995830606
832
+ ],
833
+ "mean": 59.18103711637262,
834
+ "median": 59.18022814006845,
835
+ "std": 0.11931558432079407,
836
+ "min": 59.016256918399,
837
+ "max": 59.381548995830606,
838
+ "p25": 59.12349419850713,
839
+ "p75": 59.20365732905793,
840
+ "p90": 59.310392329121534,
841
+ "p95": 59.34597066247607,
842
+ "p99": 59.3744333291597,
843
+ "unit": "tokens/sec"
844
+ },
845
+ "time_per_output_token_seconds": {
846
+ "name": "time_per_output_token_seconds",
847
+ "measurements": [
848
+ 0.01691375,
849
+ 0.016944483642578126,
850
+ 0.016897535400390625,
851
+ 0.016890848388671876,
852
+ 0.016840247802734375
853
+ ],
854
+ "mean": 0.016897373046875004,
855
+ "median": 0.016897535400390625,
856
+ "std": 3.403969407212576e-05,
857
+ "min": 0.016840247802734375,
858
+ "max": 0.016944483642578126,
859
+ "p25": 0.016890848388671876,
860
+ "p75": 0.01691375,
861
+ "p90": 0.016932190185546876,
862
+ "p95": 0.0169383369140625,
863
+ "p99": 0.016943254296875,
864
+ "unit": "seconds/token"
865
+ }
866
+ },
867
+ "gpu_metrics": {
868
+ "gpu_monitoring_status": "disabled",
869
+ "gpu_monitoring_reason": "no_gpus_available"
870
+ }
871
+ },
872
+ {
873
+ "scenario_name": "compiled_compile_max-autotune_sdpa_math",
874
+ "metadata": {
875
+ "timestamp": "2025-10-05T16:45:11.563009",
876
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
877
+ "hardware_info": {
878
+ "gpu_name": "unknown",
879
+ "gpu_memory_total_mb": 0,
880
+ "cpu_count": 160,
881
+ "memory_total_mb": 1285717,
882
+ "python_version": "3.12.10",
883
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
884
+ "cuda_version": null
885
+ },
886
+ "config": {
887
+ "name": "compiled",
888
+ "model_id": "meta-llama/Llama-2-7b-hf",
889
+ "variant": "compiled",
890
+ "warmup_iterations": 3,
891
+ "measurement_iterations": 5,
892
+ "num_tokens_to_generate": 100,
893
+ "device": "cuda",
894
+ "torch_dtype": "float16",
895
+ "compile_mode": "max-autotune",
896
+ "compile_options": {},
897
+ "use_cache": true,
898
+ "batch_size": 1,
899
+ "sequence_length": null,
900
+ "attn_implementation": "sdpa",
901
+ "sdpa_backend": "math",
902
+ "custom_params": {}
903
+ }
904
+ },
905
+ "measurements": {
906
+ "latency_seconds": {
907
+ "name": "latency_seconds",
908
+ "measurements": [
909
+ 2.0161763916015625,
910
+ 1.992341552734375,
911
+ 1.9950142822265624,
912
+ 2.0013905029296875,
913
+ 1.98680908203125
914
+ ],
915
+ "mean": 1.9983463623046878,
916
+ "median": 1.9950142822265624,
917
+ "std": 0.010074185511521987,
918
+ "min": 1.98680908203125,
919
+ "max": 2.0161763916015625,
920
+ "p25": 1.992341552734375,
921
+ "p75": 2.0013905029296875,
922
+ "p90": 2.0102620361328123,
923
+ "p95": 2.0132192138671874,
924
+ "p99": 2.0155849560546875,
925
+ "unit": "seconds"
926
+ },
927
+ "time_to_first_token_seconds": {
928
+ "name": "time_to_first_token_seconds",
929
+ "measurements": [
930
+ 0.020864751815795897,
931
+ 0.021187063217163085,
932
+ 0.020988908767700196,
933
+ 0.02157137680053711,
934
+ 0.020903072357177734
935
+ ],
936
+ "mean": 0.021103034591674803,
937
+ "median": 0.020988908767700196,
938
+ "std": 0.0002593042671098856,
939
+ "min": 0.020864751815795897,
940
+ "max": 0.02157137680053711,
941
+ "p25": 0.020903072357177734,
942
+ "p75": 0.021187063217163085,
943
+ "p90": 0.0214176513671875,
944
+ "p95": 0.021494514083862305,
945
+ "p99": 0.02155600425720215,
946
+ "unit": "seconds"
947
+ },
948
+ "tokens_per_second": {
949
+ "name": "tokens_per_second",
950
+ "measurements": [
951
+ 49.5988349117432,
952
+ 50.19219714749999,
953
+ 50.124954438117435,
954
+ 49.965261578696115,
955
+ 50.331962393569896
956
+ ],
957
+ "mean": 50.04264209392532,
958
+ "median": 50.124954438117435,
959
+ "std": 0.2513168641122693,
960
+ "min": 49.5988349117432,
961
+ "max": 50.331962393569896,
962
+ "p25": 49.965261578696115,
963
+ "p75": 50.19219714749999,
964
+ "p90": 50.276056295141935,
965
+ "p95": 50.30400934435591,
966
+ "p99": 50.3263717837271,
967
+ "unit": "tokens/sec"
968
+ },
969
+ "time_per_output_token_seconds": {
970
+ "name": "time_per_output_token_seconds",
971
+ "measurements": [
972
+ 0.020161763916015624,
973
+ 0.019923415527343752,
974
+ 0.019950142822265624,
975
+ 0.020013905029296876,
976
+ 0.0198680908203125
977
+ ],
978
+ "mean": 0.019983463623046874,
979
+ "median": 0.019950142822265624,
980
+ "std": 0.00010074185511521958,
981
+ "min": 0.0198680908203125,
982
+ "max": 0.020161763916015624,
983
+ "p25": 0.019923415527343752,
984
+ "p75": 0.020013905029296876,
985
+ "p90": 0.020102620361328125,
986
+ "p95": 0.020132192138671873,
987
+ "p99": 0.020155849560546873,
988
+ "unit": "seconds/token"
989
+ }
990
+ },
991
+ "gpu_metrics": {
992
+ "gpu_monitoring_status": "disabled",
993
+ "gpu_monitoring_reason": "no_gpus_available"
994
+ }
995
+ },
996
+ {
997
+ "scenario_name": "compiled_compile_max-autotune_sdpa_efficient_attention",
998
+ "metadata": {
999
+ "timestamp": "2025-10-05T16:45:41.893535",
1000
+ "commit_id": "e11a00a16f925b7d3b52f5007bdce3464edb361f",
1001
+ "hardware_info": {
1002
+ "gpu_name": "unknown",
1003
+ "gpu_memory_total_mb": 0,
1004
+ "cpu_count": 160,
1005
+ "memory_total_mb": 1285717,
1006
+ "python_version": "3.12.10",
1007
+ "torch_version": "2.7.1+rocm6.4.1.git2a215e4a",
1008
+ "cuda_version": null
1009
+ },
1010
+ "config": {
1011
+ "name": "compiled",
1012
+ "model_id": "meta-llama/Llama-2-7b-hf",
1013
+ "variant": "compiled",
1014
+ "warmup_iterations": 3,
1015
+ "measurement_iterations": 5,
1016
+ "num_tokens_to_generate": 100,
1017
+ "device": "cuda",
1018
+ "torch_dtype": "float16",
1019
+ "compile_mode": "max-autotune",
1020
+ "compile_options": {},
1021
+ "use_cache": true,
1022
+ "batch_size": 1,
1023
+ "sequence_length": null,
1024
+ "attn_implementation": "sdpa",
1025
+ "sdpa_backend": "efficient_attention",
1026
+ "custom_params": {}
1027
+ }
1028
+ },
1029
+ "measurements": {
1030
+ "latency_seconds": {
1031
+ "name": "latency_seconds",
1032
+ "measurements": [
1033
+ 1.69346435546875,
1034
+ 1.6841131591796874,
1035
+ 1.6800130615234374,
1036
+ 1.6840782470703124,
1037
+ 1.6876658935546875
1038
+ ],
1039
+ "mean": 1.685866943359375,
1040
+ "median": 1.6841131591796874,
1041
+ "std": 0.004505552540531352,
1042
+ "min": 1.6800130615234374,
1043
+ "max": 1.69346435546875,
1044
+ "p25": 1.6840782470703124,
1045
+ "p75": 1.6876658935546875,
1046
+ "p90": 1.691144970703125,
1047
+ "p95": 1.6923046630859375,
1048
+ "p99": 1.6932324169921875,
1049
+ "unit": "seconds"
1050
+ },
1051
+ "time_to_first_token_seconds": {
1052
+ "name": "time_to_first_token_seconds",
1053
+ "measurements": [
1054
+ 0.017702438354492187,
1055
+ 0.01763976287841797,
1056
+ 0.017580963134765624,
1057
+ 0.01745112609863281,
1058
+ 0.017634279251098634
1059
+ ],
1060
+ "mean": 0.017601713943481444,
1061
+ "median": 0.017634279251098634,
1062
+ "std": 8.457058916902933e-05,
1063
+ "min": 0.01745112609863281,
1064
+ "max": 0.017702438354492187,
1065
+ "p25": 0.017580963134765624,
1066
+ "p75": 0.01763976287841797,
1067
+ "p90": 0.0176773681640625,
1068
+ "p95": 0.017689903259277343,
1069
+ "p99": 0.01769993133544922,
1070
+ "unit": "seconds"
1071
+ },
1072
+ "tokens_per_second": {
1073
+ "name": "tokens_per_second",
1074
+ "measurements": [
1075
+ 59.05054905765646,
1076
+ 59.37843277034239,
1077
+ 59.52334674667345,
1078
+ 59.379663726411685,
1079
+ 59.253434214619666
1080
+ ],
1081
+ "mean": 59.31708530314073,
1082
+ "median": 59.37843277034239,
1083
+ "std": 0.15831385606200407,
1084
+ "min": 59.05054905765646,
1085
+ "max": 59.52334674667345,
1086
+ "p25": 59.253434214619666,
1087
+ "p75": 59.379663726411685,
1088
+ "p90": 59.465873538568744,
1089
+ "p95": 59.49461014262109,
1090
+ "p99": 59.517599425862976,
1091
+ "unit": "tokens/sec"
1092
+ },
1093
+ "time_per_output_token_seconds": {
1094
+ "name": "time_per_output_token_seconds",
1095
+ "measurements": [
1096
+ 0.0169346435546875,
1097
+ 0.016841131591796874,
1098
+ 0.016800130615234374,
1099
+ 0.016840782470703124,
1100
+ 0.016876658935546875
1101
+ ],
1102
+ "mean": 0.01685866943359375,
1103
+ "median": 0.016841131591796874,
1104
+ "std": 4.505552540531338e-05,
1105
+ "min": 0.016800130615234374,
1106
+ "max": 0.0169346435546875,
1107
+ "p25": 0.016840782470703124,
1108
+ "p75": 0.016876658935546875,
1109
+ "p90": 0.01691144970703125,
1110
+ "p95": 0.016923046630859376,
1111
+ "p99": 0.016932324169921874,
1112
+ "unit": "seconds/token"
1113
+ }
1114
+ },
1115
+ "gpu_metrics": {
1116
+ "gpu_monitoring_status": "disabled",
1117
+ "gpu_monitoring_reason": "no_gpus_available"
1118
+ }
1119
+ }
1120
+ ]
1121
+ }