hf-transformers-bot commited on
Commit
3da46f7
·
verified ·
1 Parent(s): 9a04733

Upload benchmark results for run 18228176649

Browse files
2025-10-03/18228176649/benchmark_results/Llama-2-7b-hf/Llama-2-7b-hf_benchmark_20251003_164807.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": "2025-10-03T16:44:52.341348",
8
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.3160628662109375,
42
+ 1.37740625,
43
+ 1.303656005859375,
44
+ 1.4001142578125,
45
+ 1.451359375
46
+ ],
47
+ "mean": 1.3697197509765624,
48
+ "median": 1.37740625,
49
+ "std": 0.05457364973205601,
50
+ "min": 1.303656005859375,
51
+ "max": 1.451359375,
52
+ "p25": 1.3160628662109375,
53
+ "p75": 1.4001142578125,
54
+ "p90": 1.430861328125,
55
+ "p95": 1.4411103515625001,
56
+ "p99": 1.4493095703125,
57
+ "unit": "seconds"
58
+ },
59
+ "time_to_first_token_seconds": {
60
+ "name": "time_to_first_token_seconds",
61
+ "measurements": [
62
+ 0.013936227798461914,
63
+ 0.01556003189086914,
64
+ 0.015767666816711427,
65
+ 0.01387426471710205,
66
+ 0.015369832038879394
67
+ ],
68
+ "mean": 0.014901604652404785,
69
+ "median": 0.015369832038879394,
70
+ "std": 0.0008234326686544091,
71
+ "min": 0.01387426471710205,
72
+ "max": 0.015767666816711427,
73
+ "p25": 0.013936227798461914,
74
+ "p75": 0.01556003189086914,
75
+ "p90": 0.015684612846374513,
76
+ "p95": 0.01572613983154297,
77
+ "p99": 0.015759361419677737,
78
+ "unit": "seconds"
79
+ },
80
+ "tokens_per_second": {
81
+ "name": "tokens_per_second",
82
+ "measurements": [
83
+ 75.98421212803378,
84
+ 72.60022233818091,
85
+ 76.70735190153144,
86
+ 71.42274242406276,
87
+ 68.90092262641704
88
+ ],
89
+ "mean": 73.12309028364518,
90
+ "median": 72.60022233818091,
91
+ "std": 2.899111002069914,
92
+ "min": 68.90092262641704,
93
+ "max": 76.70735190153144,
94
+ "p25": 71.42274242406276,
95
+ "p75": 75.98421212803378,
96
+ "p90": 76.41809599213238,
97
+ "p95": 76.56272394683191,
98
+ "p99": 76.67842631059153,
99
+ "unit": "tokens/sec"
100
+ },
101
+ "time_per_output_token_seconds": {
102
+ "name": "time_per_output_token_seconds",
103
+ "measurements": [
104
+ 0.013160628662109375,
105
+ 0.0137740625,
106
+ 0.013036560058593749,
107
+ 0.014001142578125,
108
+ 0.01451359375
109
+ ],
110
+ "mean": 0.013697197509765623,
111
+ "median": 0.0137740625,
112
+ "std": 0.00054573649732056,
113
+ "min": 0.013036560058593749,
114
+ "max": 0.01451359375,
115
+ "p25": 0.013160628662109375,
116
+ "p75": 0.014001142578125,
117
+ "p90": 0.01430861328125,
118
+ "p95": 0.014411103515624999,
119
+ "p99": 0.014493095703125,
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": "2025-10-03T16:45:09.005957",
132
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 3.11150244140625,
166
+ 3.05658154296875,
167
+ 3.07338720703125,
168
+ 3.160876220703125,
169
+ 3.154896728515625
170
+ ],
171
+ "mean": 3.111448828125,
172
+ "median": 3.11150244140625,
173
+ "std": 0.04192826264111398,
174
+ "min": 3.05658154296875,
175
+ "max": 3.160876220703125,
176
+ "p25": 3.07338720703125,
177
+ "p75": 3.154896728515625,
178
+ "p90": 3.158484423828125,
179
+ "p95": 3.159680322265625,
180
+ "p99": 3.160637041015625,
181
+ "unit": "seconds"
182
+ },
183
+ "time_to_first_token_seconds": {
184
+ "name": "time_to_first_token_seconds",
185
+ "measurements": [
186
+ 0.01656410789489746,
187
+ 0.016609506607055666,
188
+ 0.016440469741821288,
189
+ 0.016609222412109374,
190
+ 0.01530141258239746
191
+ ],
192
+ "mean": 0.016304943847656246,
193
+ "median": 0.01656410789489746,
194
+ "std": 0.0005055595298127066,
195
+ "min": 0.01530141258239746,
196
+ "max": 0.016609506607055666,
197
+ "p25": 0.016440469741821288,
198
+ "p75": 0.016609222412109374,
199
+ "p90": 0.01660939292907715,
200
+ "p95": 0.016609449768066407,
201
+ "p99": 0.016609495239257813,
202
+ "unit": "seconds"
203
+ },
204
+ "tokens_per_second": {
205
+ "name": "tokens_per_second",
206
+ "measurements": [
207
+ 32.13881457049566,
208
+ 32.71628732759851,
209
+ 32.537390593421314,
210
+ 31.636797209906362,
211
+ 31.696758596294806
212
+ ],
213
+ "mean": 32.14520965954334,
214
+ "median": 32.13881457049566,
215
+ "std": 0.43349003681451986,
216
+ "min": 31.636797209906362,
217
+ "max": 32.71628732759851,
218
+ "p25": 31.696758596294806,
219
+ "p75": 32.537390593421314,
220
+ "p90": 32.64472863392763,
221
+ "p95": 32.68050798076307,
222
+ "p99": 32.70913145823142,
223
+ "unit": "tokens/sec"
224
+ },
225
+ "time_per_output_token_seconds": {
226
+ "name": "time_per_output_token_seconds",
227
+ "measurements": [
228
+ 0.0311150244140625,
229
+ 0.0305658154296875,
230
+ 0.030733872070312498,
231
+ 0.03160876220703125,
232
+ 0.03154896728515625
233
+ ],
234
+ "mean": 0.03111448828125,
235
+ "median": 0.0311150244140625,
236
+ "std": 0.0004192826264111396,
237
+ "min": 0.0305658154296875,
238
+ "max": 0.03160876220703125,
239
+ "p25": 0.030733872070312498,
240
+ "p75": 0.03154896728515625,
241
+ "p90": 0.03158484423828125,
242
+ "p95": 0.031596803222656254,
243
+ "p99": 0.03160637041015625,
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": "2025-10-03T16:45:51.865105",
256
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.4151429443359376,
290
+ 1.41451318359375,
291
+ 1.4095068359375,
292
+ 1.3999083251953126,
293
+ 1.4039166259765625
294
+ ],
295
+ "mean": 1.4085975830078126,
296
+ "median": 1.4095068359375,
297
+ "std": 0.005934262989768861,
298
+ "min": 1.3999083251953126,
299
+ "max": 1.4151429443359376,
300
+ "p25": 1.4039166259765625,
301
+ "p75": 1.41451318359375,
302
+ "p90": 1.4148910400390624,
303
+ "p95": 1.4150169921875,
304
+ "p99": 1.41511775390625,
305
+ "unit": "seconds"
306
+ },
307
+ "time_to_first_token_seconds": {
308
+ "name": "time_to_first_token_seconds",
309
+ "measurements": [
310
+ 0.015233690261840821,
311
+ 0.015303487777709962,
312
+ 0.015150811195373535,
313
+ 0.015429723739624023,
314
+ 0.015130770683288575
315
+ ],
316
+ "mean": 0.015249696731567383,
317
+ "median": 0.015233690261840821,
318
+ "std": 0.0001090698506239043,
319
+ "min": 0.015130770683288575,
320
+ "max": 0.015429723739624023,
321
+ "p25": 0.015150811195373535,
322
+ "p75": 0.015303487777709962,
323
+ "p90": 0.015379229354858399,
324
+ "p95": 0.015404476547241211,
325
+ "p99": 0.01542467430114746,
326
+ "unit": "seconds"
327
+ },
328
+ "tokens_per_second": {
329
+ "name": "tokens_per_second",
330
+ "measurements": [
331
+ 70.66423953866051,
332
+ 70.69570023090017,
333
+ 70.94680029237841,
334
+ 71.43324902082298,
335
+ 71.22930104944098
336
+ ],
337
+ "mean": 70.99385802644062,
338
+ "median": 70.94680029237841,
339
+ "std": 0.2994190080428845,
340
+ "min": 70.66423953866051,
341
+ "max": 71.43324902082298,
342
+ "p25": 70.69570023090017,
343
+ "p75": 71.22930104944098,
344
+ "p90": 71.35166983227018,
345
+ "p95": 71.39245942654658,
346
+ "p99": 71.42509110196771,
347
+ "unit": "tokens/sec"
348
+ },
349
+ "time_per_output_token_seconds": {
350
+ "name": "time_per_output_token_seconds",
351
+ "measurements": [
352
+ 0.014151429443359375,
353
+ 0.014145131835937498,
354
+ 0.014095068359375,
355
+ 0.013999083251953126,
356
+ 0.014039166259765624
357
+ ],
358
+ "mean": 0.014085975830078127,
359
+ "median": 0.014095068359375,
360
+ "std": 5.934262989768861e-05,
361
+ "min": 0.013999083251953126,
362
+ "max": 0.014151429443359375,
363
+ "p25": 0.014039166259765624,
364
+ "p75": 0.014145131835937498,
365
+ "p90": 0.014148910400390624,
366
+ "p95": 0.014150169921875,
367
+ "p99": 0.0141511775390625,
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": "2025-10-03T16:46:10.001656",
380
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.3208355712890625,
414
+ 1.3591400146484376,
415
+ 1.4103245849609376,
416
+ 1.4576373291015625,
417
+ 1.470456298828125
418
+ ],
419
+ "mean": 1.403678759765625,
420
+ "median": 1.4103245849609376,
421
+ "std": 0.057029163792096015,
422
+ "min": 1.3208355712890625,
423
+ "max": 1.470456298828125,
424
+ "p25": 1.3591400146484376,
425
+ "p75": 1.4576373291015625,
426
+ "p90": 1.4653287109375,
427
+ "p95": 1.4678925048828124,
428
+ "p99": 1.4699435400390624,
429
+ "unit": "seconds"
430
+ },
431
+ "time_to_first_token_seconds": {
432
+ "name": "time_to_first_token_seconds",
433
+ "measurements": [
434
+ 0.01700680351257324,
435
+ 0.017101724624633788,
436
+ 0.0154491605758667,
437
+ 0.017015764236450195,
438
+ 0.01683308792114258
439
+ ],
440
+ "mean": 0.0166813081741333,
441
+ "median": 0.01700680351257324,
442
+ "std": 0.0006222190707997363,
443
+ "min": 0.0154491605758667,
444
+ "max": 0.017101724624633788,
445
+ "p25": 0.01683308792114258,
446
+ "p75": 0.017015764236450195,
447
+ "p90": 0.01706734046936035,
448
+ "p95": 0.01708453254699707,
449
+ "p99": 0.017098286209106445,
450
+ "unit": "seconds"
451
+ },
452
+ "tokens_per_second": {
453
+ "name": "tokens_per_second",
454
+ "measurements": [
455
+ 75.70965090106222,
456
+ 73.57593693234507,
457
+ 70.90566318303935,
458
+ 68.60417060095227,
459
+ 68.00610128957567
460
+ ],
461
+ "mean": 71.36030458139491,
462
+ "median": 70.90566318303935,
463
+ "std": 2.9284635276909565,
464
+ "min": 68.00610128957567,
465
+ "max": 75.70965090106222,
466
+ "p25": 68.60417060095227,
467
+ "p75": 73.57593693234507,
468
+ "p90": 74.85616531357536,
469
+ "p95": 75.28290810731879,
470
+ "p99": 75.62430234231353,
471
+ "unit": "tokens/sec"
472
+ },
473
+ "time_per_output_token_seconds": {
474
+ "name": "time_per_output_token_seconds",
475
+ "measurements": [
476
+ 0.013208355712890625,
477
+ 0.013591400146484377,
478
+ 0.014103245849609376,
479
+ 0.014576373291015625,
480
+ 0.01470456298828125
481
+ ],
482
+ "mean": 0.01403678759765625,
483
+ "median": 0.014103245849609376,
484
+ "std": 0.0005702916379209601,
485
+ "min": 0.013208355712890625,
486
+ "max": 0.01470456298828125,
487
+ "p25": 0.013591400146484377,
488
+ "p75": 0.014576373291015625,
489
+ "p90": 0.014653287109375,
490
+ "p95": 0.014678925048828124,
491
+ "p99": 0.014699435400390624,
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": "2025-10-03T16:46:27.669274",
504
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.36246533203125,
538
+ 1.3649266357421874,
539
+ 1.36162646484375,
540
+ 1.3571387939453126,
541
+ 1.3584180908203125
542
+ ],
543
+ "mean": 1.3609150634765625,
544
+ "median": 1.36162646484375,
545
+ "std": 0.0028105949746602466,
546
+ "min": 1.3571387939453126,
547
+ "max": 1.3649266357421874,
548
+ "p25": 1.3584180908203125,
549
+ "p75": 1.36246533203125,
550
+ "p90": 1.3639421142578125,
551
+ "p95": 1.3644343749999999,
552
+ "p99": 1.36482818359375,
553
+ "unit": "seconds"
554
+ },
555
+ "time_to_first_token_seconds": {
556
+ "name": "time_to_first_token_seconds",
557
+ "measurements": [
558
+ 0.014647257804870606,
559
+ 0.014778214454650879,
560
+ 0.014912052154541016,
561
+ 0.015040847778320312,
562
+ 0.01500036907196045
563
+ ],
564
+ "mean": 0.014875748252868654,
565
+ "median": 0.014912052154541016,
566
+ "std": 0.00014548326484736546,
567
+ "min": 0.014647257804870606,
568
+ "max": 0.015040847778320312,
569
+ "p25": 0.014778214454650879,
570
+ "p75": 0.01500036907196045,
571
+ "p90": 0.015024656295776368,
572
+ "p95": 0.01503275203704834,
573
+ "p99": 0.015039228630065918,
574
+ "unit": "seconds"
575
+ },
576
+ "tokens_per_second": {
577
+ "name": "tokens_per_second",
578
+ "measurements": [
579
+ 73.39636293784712,
580
+ 73.26401095955195,
581
+ 73.4415807726499,
582
+ 73.68443113271553,
583
+ 73.61503845963408
584
+ ],
585
+ "mean": 73.48028485247971,
586
+ "median": 73.4415807726499,
587
+ "std": 0.15175607314927192,
588
+ "min": 73.26401095955195,
589
+ "max": 73.68443113271553,
590
+ "p25": 73.39636293784712,
591
+ "p75": 73.61503845963408,
592
+ "p90": 73.65667406348295,
593
+ "p95": 73.67055259809923,
594
+ "p99": 73.68165542579227,
595
+ "unit": "tokens/sec"
596
+ },
597
+ "time_per_output_token_seconds": {
598
+ "name": "time_per_output_token_seconds",
599
+ "measurements": [
600
+ 0.013624653320312501,
601
+ 0.013649266357421874,
602
+ 0.0136162646484375,
603
+ 0.013571387939453125,
604
+ 0.013584180908203125
605
+ ],
606
+ "mean": 0.013609150634765626,
607
+ "median": 0.0136162646484375,
608
+ "std": 2.8105949746602326e-05,
609
+ "min": 0.013571387939453125,
610
+ "max": 0.013649266357421874,
611
+ "p25": 0.013584180908203125,
612
+ "p75": 0.013624653320312501,
613
+ "p90": 0.013639421142578125,
614
+ "p95": 0.01364434375,
615
+ "p99": 0.013648281835937498,
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": "2025-10-03T16:46:44.619710",
628
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.3491339111328124,
662
+ 1.2943673095703125,
663
+ 1.3947305908203125,
664
+ 1.3534305419921875,
665
+ 1.3452457275390626
666
+ ],
667
+ "mean": 1.3473816162109375,
668
+ "median": 1.3491339111328124,
669
+ "std": 0.03192699223592599,
670
+ "min": 1.2943673095703125,
671
+ "max": 1.3947305908203125,
672
+ "p25": 1.3452457275390626,
673
+ "p75": 1.3534305419921875,
674
+ "p90": 1.3782105712890624,
675
+ "p95": 1.3864705810546876,
676
+ "p99": 1.3930785888671875,
677
+ "unit": "seconds"
678
+ },
679
+ "time_to_first_token_seconds": {
680
+ "name": "time_to_first_token_seconds",
681
+ "measurements": [
682
+ 0.014420102119445801,
683
+ 0.014625535964965821,
684
+ 0.014871251106262208,
685
+ 0.01577282905578613,
686
+ 0.014719014167785644
687
+ ],
688
+ "mean": 0.014881746482849123,
689
+ "median": 0.014719014167785644,
690
+ "std": 0.000468909257495253,
691
+ "min": 0.014420102119445801,
692
+ "max": 0.01577282905578613,
693
+ "p25": 0.014625535964965821,
694
+ "p75": 0.014871251106262208,
695
+ "p90": 0.015412197875976561,
696
+ "p95": 0.015592513465881346,
697
+ "p99": 0.015736765937805174,
698
+ "unit": "seconds"
699
+ },
700
+ "tokens_per_second": {
701
+ "name": "tokens_per_second",
702
+ "measurements": [
703
+ 74.12162660416274,
704
+ 77.25782261388903,
705
+ 71.69843456375678,
706
+ 73.88631843109185,
707
+ 74.33586143621203
708
+ ],
709
+ "mean": 74.26001272982249,
710
+ "median": 74.12162660416274,
711
+ "std": 1.7727435221605807,
712
+ "min": 71.69843456375678,
713
+ "max": 77.25782261388903,
714
+ "p25": 73.88631843109185,
715
+ "p75": 74.33586143621203,
716
+ "p90": 76.08903814281823,
717
+ "p95": 76.67343037835363,
718
+ "p99": 77.14094416678195,
719
+ "unit": "tokens/sec"
720
+ },
721
+ "time_per_output_token_seconds": {
722
+ "name": "time_per_output_token_seconds",
723
+ "measurements": [
724
+ 0.013491339111328125,
725
+ 0.012943673095703125,
726
+ 0.013947305908203124,
727
+ 0.013534305419921875,
728
+ 0.013452457275390626
729
+ ],
730
+ "mean": 0.013473816162109376,
731
+ "median": 0.013491339111328125,
732
+ "std": 0.00031926992235925967,
733
+ "min": 0.012943673095703125,
734
+ "max": 0.013947305908203124,
735
+ "p25": 0.013452457275390626,
736
+ "p75": 0.013534305419921875,
737
+ "p90": 0.013782105712890624,
738
+ "p95": 0.013864705810546874,
739
+ "p99": 0.013930785888671873,
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": "2025-10-03T16:47:04.101417",
752
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.327487060546875,
786
+ 1.329615966796875,
787
+ 1.3253076171875,
788
+ 1.3936593017578125,
789
+ 1.6801934814453126
790
+ ],
791
+ "mean": 1.411252685546875,
792
+ "median": 1.329615966796875,
793
+ "std": 0.13689884397381824,
794
+ "min": 1.3253076171875,
795
+ "max": 1.6801934814453126,
796
+ "p25": 1.327487060546875,
797
+ "p75": 1.3936593017578125,
798
+ "p90": 1.5655798095703126,
799
+ "p95": 1.6228866455078126,
800
+ "p99": 1.6687321142578126,
801
+ "unit": "seconds"
802
+ },
803
+ "time_to_first_token_seconds": {
804
+ "name": "time_to_first_token_seconds",
805
+ "measurements": [
806
+ 0.014029271125793457,
807
+ 0.014088987350463867,
808
+ 0.014026227951049805,
809
+ 0.01392655086517334,
810
+ 0.017052675247192384
811
+ ],
812
+ "mean": 0.01462474250793457,
813
+ "median": 0.014029271125793457,
814
+ "std": 0.0012150858724296032,
815
+ "min": 0.01392655086517334,
816
+ "max": 0.017052675247192384,
817
+ "p25": 0.014026227951049805,
818
+ "p75": 0.014088987350463867,
819
+ "p90": 0.01586720008850098,
820
+ "p95": 0.01645993766784668,
821
+ "p99": 0.016934127731323245,
822
+ "unit": "seconds"
823
+ },
824
+ "tokens_per_second": {
825
+ "name": "tokens_per_second",
826
+ "measurements": [
827
+ 75.3303011170623,
828
+ 75.20968647880036,
829
+ 75.45418037528138,
830
+ 71.75354828391036,
831
+ 59.51695510327739
832
+ ],
833
+ "mean": 71.45293427166636,
834
+ "median": 75.20968647880036,
835
+ "std": 6.127235896778403,
836
+ "min": 59.51695510327739,
837
+ "max": 75.45418037528138,
838
+ "p25": 71.75354828391036,
839
+ "p75": 75.3303011170623,
840
+ "p90": 75.40462867199375,
841
+ "p95": 75.42940452363757,
842
+ "p99": 75.44922520495263,
843
+ "unit": "tokens/sec"
844
+ },
845
+ "time_per_output_token_seconds": {
846
+ "name": "time_per_output_token_seconds",
847
+ "measurements": [
848
+ 0.01327487060546875,
849
+ 0.01329615966796875,
850
+ 0.013253076171875,
851
+ 0.013936593017578125,
852
+ 0.016801934814453124
853
+ ],
854
+ "mean": 0.014112526855468748,
855
+ "median": 0.01329615966796875,
856
+ "std": 0.0013689884397381819,
857
+ "min": 0.013253076171875,
858
+ "max": 0.016801934814453124,
859
+ "p25": 0.01327487060546875,
860
+ "p75": 0.013936593017578125,
861
+ "p90": 0.015655798095703125,
862
+ "p95": 0.016228866455078123,
863
+ "p99": 0.016687321142578124,
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": "2025-10-03T16:47:22.654655",
876
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.9599453125,
910
+ 1.9498074951171875,
911
+ 2.05333056640625,
912
+ 1.969304931640625,
913
+ 1.9931611328125
914
+ ],
915
+ "mean": 1.9851098876953124,
916
+ "median": 1.969304931640625,
917
+ "std": 0.03700863400372719,
918
+ "min": 1.9498074951171875,
919
+ "max": 2.05333056640625,
920
+ "p25": 1.9599453125,
921
+ "p75": 1.9931611328125,
922
+ "p90": 2.02926279296875,
923
+ "p95": 2.0412966796874996,
924
+ "p99": 2.0509237890625,
925
+ "unit": "seconds"
926
+ },
927
+ "time_to_first_token_seconds": {
928
+ "name": "time_to_first_token_seconds",
929
+ "measurements": [
930
+ 0.020356117248535158,
931
+ 0.020744667053222657,
932
+ 0.02045463562011719,
933
+ 0.02081890869140625,
934
+ 0.02059163475036621
935
+ ],
936
+ "mean": 0.020593192672729493,
937
+ "median": 0.02059163475036621,
938
+ "std": 0.00017279898711548985,
939
+ "min": 0.020356117248535158,
940
+ "max": 0.02081890869140625,
941
+ "p25": 0.02045463562011719,
942
+ "p75": 0.020744667053222657,
943
+ "p90": 0.020789212036132813,
944
+ "p95": 0.02080406036376953,
945
+ "p99": 0.020815939025878905,
946
+ "unit": "seconds"
947
+ },
948
+ "tokens_per_second": {
949
+ "name": "tokens_per_second",
950
+ "measurements": [
951
+ 51.02183176348192,
952
+ 51.28711436920074,
953
+ 48.701364327820116,
954
+ 50.779337619740865,
955
+ 50.17155831194264
956
+ ],
957
+ "mean": 50.392241278437254,
958
+ "median": 50.779337619740865,
959
+ "std": 0.9224740038300465,
960
+ "min": 48.701364327820116,
961
+ "max": 51.28711436920074,
962
+ "p25": 50.17155831194264,
963
+ "p75": 51.02183176348192,
964
+ "p90": 51.18100132691321,
965
+ "p95": 51.23405784805697,
966
+ "p99": 51.27650306497198,
967
+ "unit": "tokens/sec"
968
+ },
969
+ "time_per_output_token_seconds": {
970
+ "name": "time_per_output_token_seconds",
971
+ "measurements": [
972
+ 0.019599453125,
973
+ 0.019498074951171875,
974
+ 0.020533305664062498,
975
+ 0.01969304931640625,
976
+ 0.019931611328125
977
+ ],
978
+ "mean": 0.019851098876953126,
979
+ "median": 0.01969304931640625,
980
+ "std": 0.00037008634003727215,
981
+ "min": 0.019498074951171875,
982
+ "max": 0.020533305664062498,
983
+ "p25": 0.019599453125,
984
+ "p75": 0.019931611328125,
985
+ "p90": 0.020292627929687498,
986
+ "p95": 0.020412966796874998,
987
+ "p99": 0.020509237890624997,
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": "2025-10-03T16:47:53.671023",
1000
+ "commit_id": "1bc75db9bdb54c3fe1042c298690cf37132fe2bc",
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
+ 1.340429931640625,
1034
+ 1.382703125,
1035
+ 1.3324952392578124,
1036
+ 1.6198009033203125,
1037
+ 1.3458236083984374
1038
+ ],
1039
+ "mean": 1.4042505615234373,
1040
+ "median": 1.3458236083984374,
1041
+ "std": 0.10914380403214358,
1042
+ "min": 1.3324952392578124,
1043
+ "max": 1.6198009033203125,
1044
+ "p25": 1.340429931640625,
1045
+ "p75": 1.382703125,
1046
+ "p90": 1.5249617919921874,
1047
+ "p95": 1.5723813476562498,
1048
+ "p99": 1.6103169921875,
1049
+ "unit": "seconds"
1050
+ },
1051
+ "time_to_first_token_seconds": {
1052
+ "name": "time_to_first_token_seconds",
1053
+ "measurements": [
1054
+ 0.014092945098876953,
1055
+ 0.014229660987854003,
1056
+ 0.014131665229797363,
1057
+ 0.014085386276245117,
1058
+ 0.014453974723815918
1059
+ ],
1060
+ "mean": 0.01419872646331787,
1061
+ "median": 0.014131665229797363,
1062
+ "std": 0.0001375801754654477,
1063
+ "min": 0.014085386276245117,
1064
+ "max": 0.014453974723815918,
1065
+ "p25": 0.014092945098876953,
1066
+ "p75": 0.014229660987854003,
1067
+ "p90": 0.014364249229431152,
1068
+ "p95": 0.014409111976623536,
1069
+ "p99": 0.014445002174377441,
1070
+ "unit": "seconds"
1071
+ },
1072
+ "tokens_per_second": {
1073
+ "name": "tokens_per_second",
1074
+ "measurements": [
1075
+ 74.60292973136207,
1076
+ 72.32210457324308,
1077
+ 75.04717244295678,
1078
+ 61.73598236364559,
1079
+ 74.30394248990952
1080
+ ],
1081
+ "mean": 71.6024263202234,
1082
+ "median": 74.30394248990952,
1083
+ "std": 5.020600348993678,
1084
+ "min": 61.73598236364559,
1085
+ "max": 75.04717244295678,
1086
+ "p25": 72.32210457324308,
1087
+ "p75": 74.60292973136207,
1088
+ "p90": 74.8694753583189,
1089
+ "p95": 74.95832390063784,
1090
+ "p99": 75.02940273449299,
1091
+ "unit": "tokens/sec"
1092
+ },
1093
+ "time_per_output_token_seconds": {
1094
+ "name": "time_per_output_token_seconds",
1095
+ "measurements": [
1096
+ 0.01340429931640625,
1097
+ 0.01382703125,
1098
+ 0.013324952392578124,
1099
+ 0.016198009033203124,
1100
+ 0.013458236083984373
1101
+ ],
1102
+ "mean": 0.014042505615234374,
1103
+ "median": 0.013458236083984373,
1104
+ "std": 0.0010914380403214353,
1105
+ "min": 0.013324952392578124,
1106
+ "max": 0.016198009033203124,
1107
+ "p25": 0.01340429931640625,
1108
+ "p75": 0.01382703125,
1109
+ "p90": 0.015249617919921875,
1110
+ "p95": 0.015723813476562497,
1111
+ "p99": 0.016103169921874997,
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
+ }