rtr-bench 0.2.0 → 0.2.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +3 -1
- package/dist/cli.js +114 -22
- package/docs/development-pilot-2026-11-14.json +262 -0
- package/docs/distribution.md +1 -1
- package/docs/final-run-plan-8000.json +900 -0
- package/docs/final-run-plan-8000.md +73 -0
- package/docs/learn-room.md +7 -5
- package/docs/npm-release.md +5 -5
- package/docs/preflight-audit.md +81 -0
- package/docs/preflight-providers-2026-11-14.json +497 -0
- package/package.json +7 -2
|
@@ -0,0 +1,900 @@
|
|
|
1
|
+
{
|
|
2
|
+
"createdAt": "2026-11-14T23:37:30.165Z",
|
|
3
|
+
"status": "prepared; execution budget not yet approved",
|
|
4
|
+
"packageVersion": "0.2.1",
|
|
5
|
+
"runnerVersion": "learn-room-runner-v4",
|
|
6
|
+
"basisCommit": "2414723bdd45fb38221c84b4aaa563c0f4fa135a",
|
|
7
|
+
"protocol": {
|
|
8
|
+
"partition": "evaluation",
|
|
9
|
+
"targets": [
|
|
10
|
+
130,
|
|
11
|
+
131,
|
|
12
|
+
132,
|
|
13
|
+
133,
|
|
14
|
+
134,
|
|
15
|
+
135,
|
|
16
|
+
136,
|
|
17
|
+
137,
|
|
18
|
+
138,
|
|
19
|
+
139,
|
|
20
|
+
140,
|
|
21
|
+
141,
|
|
22
|
+
142,
|
|
23
|
+
143,
|
|
24
|
+
144,
|
|
25
|
+
145,
|
|
26
|
+
146,
|
|
27
|
+
147,
|
|
28
|
+
148,
|
|
29
|
+
149,
|
|
30
|
+
150,
|
|
31
|
+
151,
|
|
32
|
+
152,
|
|
33
|
+
153,
|
|
34
|
+
154,
|
|
35
|
+
155,
|
|
36
|
+
156,
|
|
37
|
+
157,
|
|
38
|
+
158,
|
|
39
|
+
159,
|
|
40
|
+
160,
|
|
41
|
+
161,
|
|
42
|
+
162,
|
|
43
|
+
163,
|
|
44
|
+
164,
|
|
45
|
+
165,
|
|
46
|
+
166,
|
|
47
|
+
167,
|
|
48
|
+
168,
|
|
49
|
+
169,
|
|
50
|
+
170,
|
|
51
|
+
171,
|
|
52
|
+
172,
|
|
53
|
+
173,
|
|
54
|
+
174
|
|
55
|
+
],
|
|
56
|
+
"reference": [
|
|
57
|
+
1,
|
|
58
|
+
2,
|
|
59
|
+
3,
|
|
60
|
+
4,
|
|
61
|
+
5,
|
|
62
|
+
6,
|
|
63
|
+
7,
|
|
64
|
+
8,
|
|
65
|
+
9,
|
|
66
|
+
10,
|
|
67
|
+
11,
|
|
68
|
+
12,
|
|
69
|
+
13,
|
|
70
|
+
14,
|
|
71
|
+
15,
|
|
72
|
+
16,
|
|
73
|
+
17,
|
|
74
|
+
18,
|
|
75
|
+
19,
|
|
76
|
+
20,
|
|
77
|
+
21,
|
|
78
|
+
22,
|
|
79
|
+
23,
|
|
80
|
+
24,
|
|
81
|
+
25,
|
|
82
|
+
28,
|
|
83
|
+
29,
|
|
84
|
+
30,
|
|
85
|
+
31,
|
|
86
|
+
35,
|
|
87
|
+
36,
|
|
88
|
+
37,
|
|
89
|
+
38,
|
|
90
|
+
39,
|
|
91
|
+
42,
|
|
92
|
+
43,
|
|
93
|
+
44,
|
|
94
|
+
45,
|
|
95
|
+
46,
|
|
96
|
+
47,
|
|
97
|
+
48,
|
|
98
|
+
49,
|
|
99
|
+
50,
|
|
100
|
+
51,
|
|
101
|
+
52,
|
|
102
|
+
53,
|
|
103
|
+
54,
|
|
104
|
+
55,
|
|
105
|
+
56,
|
|
106
|
+
57,
|
|
107
|
+
58,
|
|
108
|
+
59,
|
|
109
|
+
60,
|
|
110
|
+
61,
|
|
111
|
+
62,
|
|
112
|
+
63,
|
|
113
|
+
64,
|
|
114
|
+
65,
|
|
115
|
+
66,
|
|
116
|
+
67,
|
|
117
|
+
68,
|
|
118
|
+
69,
|
|
119
|
+
70,
|
|
120
|
+
71,
|
|
121
|
+
72,
|
|
122
|
+
73,
|
|
123
|
+
74,
|
|
124
|
+
77,
|
|
125
|
+
78,
|
|
126
|
+
79,
|
|
127
|
+
80,
|
|
128
|
+
81,
|
|
129
|
+
82,
|
|
130
|
+
83,
|
|
131
|
+
84,
|
|
132
|
+
85,
|
|
133
|
+
86,
|
|
134
|
+
87,
|
|
135
|
+
88,
|
|
136
|
+
89,
|
|
137
|
+
90,
|
|
138
|
+
91,
|
|
139
|
+
92,
|
|
140
|
+
93,
|
|
141
|
+
94,
|
|
142
|
+
95,
|
|
143
|
+
96,
|
|
144
|
+
97,
|
|
145
|
+
98,
|
|
146
|
+
99
|
|
147
|
+
],
|
|
148
|
+
"datasetHash": "ec9d7582b5610a24f4e14dbbb05d13c636b692ec9e133b9381db0fdf483ffb2b",
|
|
149
|
+
"splitHash": "a165c72f7557f8128253148eb28e6906b39e80aec12e45c679fddde62fa31421",
|
|
150
|
+
"strategies": [
|
|
151
|
+
"none",
|
|
152
|
+
"random",
|
|
153
|
+
"similar",
|
|
154
|
+
"agent"
|
|
155
|
+
],
|
|
156
|
+
"budget": 3,
|
|
157
|
+
"seed": 7,
|
|
158
|
+
"maxModelCalls": 8,
|
|
159
|
+
"maxActions": 8,
|
|
160
|
+
"maxTokens": 8000,
|
|
161
|
+
"maxAttempts": 3,
|
|
162
|
+
"timeoutMs": 60000,
|
|
163
|
+
"agentPolicy": "transfer-check-v1",
|
|
164
|
+
"temperature": "0 where supported; provider default otherwise",
|
|
165
|
+
"reasoning": "provider defaults, unchanged from development"
|
|
166
|
+
},
|
|
167
|
+
"output": "results/final-8000",
|
|
168
|
+
"estimation": "Reprice saved prompt/completion tokens at current base rates with no cache discount; scale 3-target pilots by 15. Use the 8000-token follow-up for MiniMax/DeepSeek and the existing 45-target 4000-token final run for GLM Flash. Reasoning is included in completion tokens, not added again. Caps round up 3x estimated spend plus one conservative reservation to the next $0.25. Estimates are not guarantees.",
|
|
169
|
+
"models": [
|
|
170
|
+
{
|
|
171
|
+
"model": "z-ai/glm-5.3-flash",
|
|
172
|
+
"provider": "relace/fp4",
|
|
173
|
+
"quantization": "fp4",
|
|
174
|
+
"catalogStatus": 0,
|
|
175
|
+
"source": "https://openrouter.ai/api/v1/models/z-ai/glm-5.3-flash/endpoints",
|
|
176
|
+
"endpointPricing": {
|
|
177
|
+
"prompt": "0.0000000399",
|
|
178
|
+
"completion": "0.0000000784",
|
|
179
|
+
"input_cache_read": "0.00000003",
|
|
180
|
+
"discount": 0
|
|
181
|
+
},
|
|
182
|
+
"temperatureSupported": true,
|
|
183
|
+
"maxCompletionTokens": 131072,
|
|
184
|
+
"sourceDirectory": "results/learn-room/z-ai-glm-5-3-flash-e9d0327447ef1856",
|
|
185
|
+
"sourceMaxTokens": 4000,
|
|
186
|
+
"sourceTargets": 45,
|
|
187
|
+
"promptTokens": 368365,
|
|
188
|
+
"completionTokens": 244570,
|
|
189
|
+
"estimateUsd": 0.0338720515,
|
|
190
|
+
"reservationUsd": 0.0424653824,
|
|
191
|
+
"capUsd": 0.25,
|
|
192
|
+
"estimatedMinutes": 29.556983333333335,
|
|
193
|
+
"estimatedCalls": 347,
|
|
194
|
+
"spending": {
|
|
195
|
+
"version": "reserved-context-v1",
|
|
196
|
+
"limitUsd": 0.25,
|
|
197
|
+
"providerTag": "relace/fp4",
|
|
198
|
+
"contextTokens": 1048576,
|
|
199
|
+
"inputUsdPerToken": 3.99e-8,
|
|
200
|
+
"outputUsdPerToken": 7.84e-8
|
|
201
|
+
},
|
|
202
|
+
"cliArgs": [
|
|
203
|
+
"read-the-room",
|
|
204
|
+
"z-ai/glm-5.3-flash",
|
|
205
|
+
"--controls",
|
|
206
|
+
"--provider",
|
|
207
|
+
"relace/fp4",
|
|
208
|
+
"--split",
|
|
209
|
+
"evaluation",
|
|
210
|
+
"--limit",
|
|
211
|
+
"45",
|
|
212
|
+
"--budget",
|
|
213
|
+
"3",
|
|
214
|
+
"--seed",
|
|
215
|
+
"7",
|
|
216
|
+
"--max-model-calls",
|
|
217
|
+
"8",
|
|
218
|
+
"--max-tokens",
|
|
219
|
+
"8000",
|
|
220
|
+
"--agent-policy",
|
|
221
|
+
"transfer-check-v1",
|
|
222
|
+
"--max-usd",
|
|
223
|
+
"0.25",
|
|
224
|
+
"--output",
|
|
225
|
+
"results/final-8000/z-ai__glm-5.3-flash"
|
|
226
|
+
]
|
|
227
|
+
},
|
|
228
|
+
{
|
|
229
|
+
"model": "z-ai/glm-5.3",
|
|
230
|
+
"provider": "wafer/us",
|
|
231
|
+
"quantization": "fp4",
|
|
232
|
+
"catalogStatus": 0,
|
|
233
|
+
"source": "https://openrouter.ai/api/v1/models/z-ai/glm-5.3/endpoints",
|
|
234
|
+
"endpointPricing": {
|
|
235
|
+
"prompt": "0.0000000699",
|
|
236
|
+
"completion": "0.0000014",
|
|
237
|
+
"input_cache_read": "0.0000000689",
|
|
238
|
+
"discount": 0
|
|
239
|
+
},
|
|
240
|
+
"temperatureSupported": true,
|
|
241
|
+
"maxCompletionTokens": 943718,
|
|
242
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/z-ai__glm-5.3/z-ai-glm-5-3-3d817a7af8b62c13",
|
|
243
|
+
"sourceMaxTokens": 4000,
|
|
244
|
+
"sourceTargets": 3,
|
|
245
|
+
"promptTokens": 24321,
|
|
246
|
+
"completionTokens": 1224,
|
|
247
|
+
"estimateUsd": 0.0512045685,
|
|
248
|
+
"reservationUsd": 0.0844954624,
|
|
249
|
+
"capUsd": 0.25,
|
|
250
|
+
"estimatedMinutes": 4.17375,
|
|
251
|
+
"estimatedCalls": 345,
|
|
252
|
+
"spending": {
|
|
253
|
+
"version": "reserved-context-v1",
|
|
254
|
+
"limitUsd": 0.25,
|
|
255
|
+
"providerTag": "wafer/us",
|
|
256
|
+
"contextTokens": 1048576,
|
|
257
|
+
"inputUsdPerToken": 6.99e-8,
|
|
258
|
+
"outputUsdPerToken": 0.0000014
|
|
259
|
+
},
|
|
260
|
+
"cliArgs": [
|
|
261
|
+
"read-the-room",
|
|
262
|
+
"z-ai/glm-5.3",
|
|
263
|
+
"--controls",
|
|
264
|
+
"--provider",
|
|
265
|
+
"wafer/us",
|
|
266
|
+
"--split",
|
|
267
|
+
"evaluation",
|
|
268
|
+
"--limit",
|
|
269
|
+
"45",
|
|
270
|
+
"--budget",
|
|
271
|
+
"3",
|
|
272
|
+
"--seed",
|
|
273
|
+
"7",
|
|
274
|
+
"--max-model-calls",
|
|
275
|
+
"8",
|
|
276
|
+
"--max-tokens",
|
|
277
|
+
"8000",
|
|
278
|
+
"--agent-policy",
|
|
279
|
+
"transfer-check-v1",
|
|
280
|
+
"--max-usd",
|
|
281
|
+
"0.25",
|
|
282
|
+
"--output",
|
|
283
|
+
"results/final-8000/z-ai__glm-5.3"
|
|
284
|
+
]
|
|
285
|
+
},
|
|
286
|
+
{
|
|
287
|
+
"model": "minimax/minimax-m3.1-flash",
|
|
288
|
+
"provider": "inferact-vllm/nvfp4",
|
|
289
|
+
"quantization": "nvfp4",
|
|
290
|
+
"catalogStatus": 0,
|
|
291
|
+
"source": "https://openrouter.ai/api/v1/models/minimax/minimax-m3.1-flash/endpoints",
|
|
292
|
+
"endpointPricing": {
|
|
293
|
+
"prompt": "0.000000075",
|
|
294
|
+
"completion": "0.0000003",
|
|
295
|
+
"input_cache_read": "0.0000000075",
|
|
296
|
+
"discount": 0
|
|
297
|
+
},
|
|
298
|
+
"temperatureSupported": true,
|
|
299
|
+
"maxCompletionTokens": 131072,
|
|
300
|
+
"sourceDirectory": "results/development-token-check-8000-2026-11-14/minimax__minimax-m3.1-flash/minimax-minimax-m3-1-flash-c50c917b011c8ed5",
|
|
301
|
+
"sourceMaxTokens": 8000,
|
|
302
|
+
"sourceTargets": 3,
|
|
303
|
+
"promptTokens": 29144,
|
|
304
|
+
"completionTokens": 14135,
|
|
305
|
+
"estimateUsd": 0.0963945,
|
|
306
|
+
"reservationUsd": 0.0810432,
|
|
307
|
+
"capUsd": 0.5,
|
|
308
|
+
"estimatedMinutes": 19.2355,
|
|
309
|
+
"estimatedCalls": 375,
|
|
310
|
+
"spending": {
|
|
311
|
+
"version": "reserved-context-v1",
|
|
312
|
+
"limitUsd": 0.5,
|
|
313
|
+
"providerTag": "inferact-vllm/nvfp4",
|
|
314
|
+
"contextTokens": 1048576,
|
|
315
|
+
"inputUsdPerToken": 7.5e-8,
|
|
316
|
+
"outputUsdPerToken": 3e-7
|
|
317
|
+
},
|
|
318
|
+
"cliArgs": [
|
|
319
|
+
"read-the-room",
|
|
320
|
+
"minimax/minimax-m3.1-flash",
|
|
321
|
+
"--controls",
|
|
322
|
+
"--provider",
|
|
323
|
+
"inferact-vllm/nvfp4",
|
|
324
|
+
"--split",
|
|
325
|
+
"evaluation",
|
|
326
|
+
"--limit",
|
|
327
|
+
"45",
|
|
328
|
+
"--budget",
|
|
329
|
+
"3",
|
|
330
|
+
"--seed",
|
|
331
|
+
"7",
|
|
332
|
+
"--max-model-calls",
|
|
333
|
+
"8",
|
|
334
|
+
"--max-tokens",
|
|
335
|
+
"8000",
|
|
336
|
+
"--agent-policy",
|
|
337
|
+
"transfer-check-v1",
|
|
338
|
+
"--max-usd",
|
|
339
|
+
"0.5",
|
|
340
|
+
"--output",
|
|
341
|
+
"results/final-8000/minimax__minimax-m3.1-flash"
|
|
342
|
+
]
|
|
343
|
+
},
|
|
344
|
+
{
|
|
345
|
+
"model": "xiaomi/mimo-v2.6-flash",
|
|
346
|
+
"provider": "xiaomi/fp8",
|
|
347
|
+
"quantization": "fp8",
|
|
348
|
+
"catalogStatus": 0,
|
|
349
|
+
"source": "https://openrouter.ai/api/v1/models/xiaomi/mimo-v2.6-flash/endpoints",
|
|
350
|
+
"endpointPricing": {
|
|
351
|
+
"prompt": "0.00000014",
|
|
352
|
+
"completion": "0.00000028",
|
|
353
|
+
"input_cache_read": "0.0000000028",
|
|
354
|
+
"discount": 0
|
|
355
|
+
},
|
|
356
|
+
"temperatureSupported": true,
|
|
357
|
+
"maxCompletionTokens": 131072,
|
|
358
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/xiaomi__mimo-v2.6-flash/xiaomi-mimo-v2-6-flash-6daa4284045d1f6e",
|
|
359
|
+
"sourceMaxTokens": 4000,
|
|
360
|
+
"sourceTargets": 3,
|
|
361
|
+
"promptTokens": 26345,
|
|
362
|
+
"completionTokens": 3589,
|
|
363
|
+
"estimateUsd": 0.07039830000000001,
|
|
364
|
+
"reservationUsd": 0.14904064,
|
|
365
|
+
"capUsd": 0.5,
|
|
366
|
+
"estimatedMinutes": 35.5935,
|
|
367
|
+
"estimatedCalls": 345,
|
|
368
|
+
"spending": {
|
|
369
|
+
"version": "reserved-context-v1",
|
|
370
|
+
"limitUsd": 0.5,
|
|
371
|
+
"providerTag": "xiaomi/fp8",
|
|
372
|
+
"contextTokens": 1048576,
|
|
373
|
+
"inputUsdPerToken": 1.4e-7,
|
|
374
|
+
"outputUsdPerToken": 2.8e-7
|
|
375
|
+
},
|
|
376
|
+
"cliArgs": [
|
|
377
|
+
"read-the-room",
|
|
378
|
+
"xiaomi/mimo-v2.6-flash",
|
|
379
|
+
"--controls",
|
|
380
|
+
"--provider",
|
|
381
|
+
"xiaomi/fp8",
|
|
382
|
+
"--split",
|
|
383
|
+
"evaluation",
|
|
384
|
+
"--limit",
|
|
385
|
+
"45",
|
|
386
|
+
"--budget",
|
|
387
|
+
"3",
|
|
388
|
+
"--seed",
|
|
389
|
+
"7",
|
|
390
|
+
"--max-model-calls",
|
|
391
|
+
"8",
|
|
392
|
+
"--max-tokens",
|
|
393
|
+
"8000",
|
|
394
|
+
"--agent-policy",
|
|
395
|
+
"transfer-check-v1",
|
|
396
|
+
"--max-usd",
|
|
397
|
+
"0.5",
|
|
398
|
+
"--output",
|
|
399
|
+
"results/final-8000/xiaomi__mimo-v2.6-flash"
|
|
400
|
+
]
|
|
401
|
+
},
|
|
402
|
+
{
|
|
403
|
+
"model": "deepseek/deepseek-v4.1-flash-1017",
|
|
404
|
+
"provider": "sail-research/fp4/us",
|
|
405
|
+
"quantization": "fp4",
|
|
406
|
+
"catalogStatus": 0,
|
|
407
|
+
"source": "https://openrouter.ai/api/v1/models/deepseek/deepseek-v4.1-flash-1017/endpoints",
|
|
408
|
+
"endpointPricing": {
|
|
409
|
+
"prompt": "0.00000007272",
|
|
410
|
+
"completion": "0.000000108",
|
|
411
|
+
"input_cache_read": "0.000000007",
|
|
412
|
+
"discount": 0
|
|
413
|
+
},
|
|
414
|
+
"temperatureSupported": true,
|
|
415
|
+
"maxCompletionTokens": 384000,
|
|
416
|
+
"sourceDirectory": "results/development-token-check-8000-2026-11-14/deepseek__deepseek-v4.1-flash-1017/deepseek-deepseek-v4-1-flash-1017-4161bdb01b5ad77f",
|
|
417
|
+
"sourceMaxTokens": 8000,
|
|
418
|
+
"sourceTargets": 3,
|
|
419
|
+
"promptTokens": 28748,
|
|
420
|
+
"completionTokens": 42835,
|
|
421
|
+
"estimateUsd": 0.1007510184,
|
|
422
|
+
"reservationUsd": 0.07711644672000001,
|
|
423
|
+
"capUsd": 0.5,
|
|
424
|
+
"estimatedMinutes": 43.966499999999996,
|
|
425
|
+
"estimatedCalls": 360,
|
|
426
|
+
"spending": {
|
|
427
|
+
"version": "reserved-context-v1",
|
|
428
|
+
"limitUsd": 0.5,
|
|
429
|
+
"providerTag": "sail-research/fp4/us",
|
|
430
|
+
"contextTokens": 1048576,
|
|
431
|
+
"inputUsdPerToken": 7.272e-8,
|
|
432
|
+
"outputUsdPerToken": 1.08e-7
|
|
433
|
+
},
|
|
434
|
+
"cliArgs": [
|
|
435
|
+
"read-the-room",
|
|
436
|
+
"deepseek/deepseek-v4.1-flash-1017",
|
|
437
|
+
"--controls",
|
|
438
|
+
"--provider",
|
|
439
|
+
"sail-research/fp4/us",
|
|
440
|
+
"--split",
|
|
441
|
+
"evaluation",
|
|
442
|
+
"--limit",
|
|
443
|
+
"45",
|
|
444
|
+
"--budget",
|
|
445
|
+
"3",
|
|
446
|
+
"--seed",
|
|
447
|
+
"7",
|
|
448
|
+
"--max-model-calls",
|
|
449
|
+
"8",
|
|
450
|
+
"--max-tokens",
|
|
451
|
+
"8000",
|
|
452
|
+
"--agent-policy",
|
|
453
|
+
"transfer-check-v1",
|
|
454
|
+
"--max-usd",
|
|
455
|
+
"0.5",
|
|
456
|
+
"--output",
|
|
457
|
+
"results/final-8000/deepseek__deepseek-v4.1-flash-1017"
|
|
458
|
+
]
|
|
459
|
+
},
|
|
460
|
+
{
|
|
461
|
+
"model": "tencent/hy3",
|
|
462
|
+
"provider": "gmicloud/bf16",
|
|
463
|
+
"quantization": "bf16",
|
|
464
|
+
"catalogStatus": 0,
|
|
465
|
+
"source": "https://openrouter.ai/api/v1/models/tencent/hy3/endpoints",
|
|
466
|
+
"endpointPricing": {
|
|
467
|
+
"prompt": "0.00000014",
|
|
468
|
+
"completion": "0.00000058",
|
|
469
|
+
"input_cache_read": "0.000000035",
|
|
470
|
+
"discount": 0
|
|
471
|
+
},
|
|
472
|
+
"temperatureSupported": true,
|
|
473
|
+
"maxCompletionTokens": 235929,
|
|
474
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/tencent__hy3/tencent-hy3-ba68c928008ef8f1",
|
|
475
|
+
"sourceMaxTokens": 4000,
|
|
476
|
+
"sourceTargets": 3,
|
|
477
|
+
"promptTokens": 25401,
|
|
478
|
+
"completionTokens": 30132,
|
|
479
|
+
"estimateUsd": 0.31549049999999995,
|
|
480
|
+
"reservationUsd": 0.04134016,
|
|
481
|
+
"capUsd": 1,
|
|
482
|
+
"estimatedMinutes": 112.12299999999999,
|
|
483
|
+
"estimatedCalls": 345,
|
|
484
|
+
"spending": {
|
|
485
|
+
"version": "reserved-context-v1",
|
|
486
|
+
"limitUsd": 1,
|
|
487
|
+
"providerTag": "gmicloud/bf16",
|
|
488
|
+
"contextTokens": 262144,
|
|
489
|
+
"inputUsdPerToken": 1.4e-7,
|
|
490
|
+
"outputUsdPerToken": 5.8e-7
|
|
491
|
+
},
|
|
492
|
+
"cliArgs": [
|
|
493
|
+
"read-the-room",
|
|
494
|
+
"tencent/hy3",
|
|
495
|
+
"--controls",
|
|
496
|
+
"--provider",
|
|
497
|
+
"gmicloud/bf16",
|
|
498
|
+
"--split",
|
|
499
|
+
"evaluation",
|
|
500
|
+
"--limit",
|
|
501
|
+
"45",
|
|
502
|
+
"--budget",
|
|
503
|
+
"3",
|
|
504
|
+
"--seed",
|
|
505
|
+
"7",
|
|
506
|
+
"--max-model-calls",
|
|
507
|
+
"8",
|
|
508
|
+
"--max-tokens",
|
|
509
|
+
"8000",
|
|
510
|
+
"--agent-policy",
|
|
511
|
+
"transfer-check-v1",
|
|
512
|
+
"--max-usd",
|
|
513
|
+
"1",
|
|
514
|
+
"--output",
|
|
515
|
+
"results/final-8000/tencent__hy3"
|
|
516
|
+
]
|
|
517
|
+
},
|
|
518
|
+
{
|
|
519
|
+
"model": "google/gemini-3.8-flash",
|
|
520
|
+
"provider": "google-ai-studio",
|
|
521
|
+
"quantization": "unknown",
|
|
522
|
+
"catalogStatus": 0,
|
|
523
|
+
"source": "https://openrouter.ai/api/v1/models/google/gemini-3.8-flash/endpoints",
|
|
524
|
+
"endpointPricing": {
|
|
525
|
+
"prompt": "0.00000075",
|
|
526
|
+
"completion": "0.00000375",
|
|
527
|
+
"image": "0.00000075",
|
|
528
|
+
"audio": "0.00000075",
|
|
529
|
+
"input_audio_cache": "0.000000075",
|
|
530
|
+
"web_search": "0.014",
|
|
531
|
+
"internal_reasoning": "0.00000375",
|
|
532
|
+
"input_cache_read": "0.000000075",
|
|
533
|
+
"input_cache_write": "0.0000000416666666666667",
|
|
534
|
+
"discount": 0.5
|
|
535
|
+
},
|
|
536
|
+
"temperatureSupported": true,
|
|
537
|
+
"maxCompletionTokens": 65536,
|
|
538
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/google__gemini-3.8-flash/google-gemini-3-8-flash-504a2e4b9410a1b1",
|
|
539
|
+
"sourceMaxTokens": 4000,
|
|
540
|
+
"sourceTargets": 3,
|
|
541
|
+
"promptTokens": 25578,
|
|
542
|
+
"completionTokens": 9966,
|
|
543
|
+
"estimateUsd": 0.84834,
|
|
544
|
+
"reservationUsd": 0.816432,
|
|
545
|
+
"capUsd": 3.5,
|
|
546
|
+
"estimatedMinutes": 27.441,
|
|
547
|
+
"estimatedCalls": 360,
|
|
548
|
+
"spending": {
|
|
549
|
+
"version": "reserved-text-v2",
|
|
550
|
+
"limitUsd": 3.5,
|
|
551
|
+
"providerTag": "google-ai-studio",
|
|
552
|
+
"contextTokens": 1048576,
|
|
553
|
+
"inputUsdPerToken": 7.5e-7,
|
|
554
|
+
"outputUsdPerToken": 0.00000375,
|
|
555
|
+
"model": "google/gemini-3.8-flash",
|
|
556
|
+
"temperatureSupported": true,
|
|
557
|
+
"maxCompletionTokens": 65536,
|
|
558
|
+
"pricing": {
|
|
559
|
+
"audio": "0.00000075",
|
|
560
|
+
"completion": "0.00000375",
|
|
561
|
+
"discount": 0.5,
|
|
562
|
+
"image": "0.00000075",
|
|
563
|
+
"input_audio_cache": "0.000000075",
|
|
564
|
+
"input_cache_read": "0.000000075",
|
|
565
|
+
"input_cache_write": "0.0000000416666666666667",
|
|
566
|
+
"internal_reasoning": "0.00000375",
|
|
567
|
+
"prompt": "0.00000075",
|
|
568
|
+
"web_search": "0.014"
|
|
569
|
+
}
|
|
570
|
+
},
|
|
571
|
+
"cliArgs": [
|
|
572
|
+
"read-the-room",
|
|
573
|
+
"google/gemini-3.8-flash",
|
|
574
|
+
"--controls",
|
|
575
|
+
"--provider",
|
|
576
|
+
"google-ai-studio",
|
|
577
|
+
"--split",
|
|
578
|
+
"evaluation",
|
|
579
|
+
"--limit",
|
|
580
|
+
"45",
|
|
581
|
+
"--budget",
|
|
582
|
+
"3",
|
|
583
|
+
"--seed",
|
|
584
|
+
"7",
|
|
585
|
+
"--max-model-calls",
|
|
586
|
+
"8",
|
|
587
|
+
"--max-tokens",
|
|
588
|
+
"8000",
|
|
589
|
+
"--agent-policy",
|
|
590
|
+
"transfer-check-v1",
|
|
591
|
+
"--max-usd",
|
|
592
|
+
"3.5",
|
|
593
|
+
"--output",
|
|
594
|
+
"results/final-8000/google__gemini-3.8-flash"
|
|
595
|
+
]
|
|
596
|
+
},
|
|
597
|
+
{
|
|
598
|
+
"model": "openai/gpt-6-luna",
|
|
599
|
+
"provider": "openai",
|
|
600
|
+
"quantization": "unknown",
|
|
601
|
+
"catalogStatus": 0,
|
|
602
|
+
"source": "https://openrouter.ai/api/v1/models/openai/gpt-6-luna/endpoints",
|
|
603
|
+
"endpointPricing": {
|
|
604
|
+
"prompt": "0.0000001",
|
|
605
|
+
"completion": "0.0000005",
|
|
606
|
+
"web_search": "0.01",
|
|
607
|
+
"input_cache_read": "0.00000001",
|
|
608
|
+
"input_cache_write": "0.000000125",
|
|
609
|
+
"discount": 0,
|
|
610
|
+
"overrides": [
|
|
611
|
+
{
|
|
612
|
+
"min_prompt_tokens": 272000,
|
|
613
|
+
"prompt": "0.0000002",
|
|
614
|
+
"completion": "0.00000075",
|
|
615
|
+
"input_cache_read": "0.00000002",
|
|
616
|
+
"input_cache_write": "0.00000025"
|
|
617
|
+
}
|
|
618
|
+
]
|
|
619
|
+
},
|
|
620
|
+
"temperatureSupported": false,
|
|
621
|
+
"maxCompletionTokens": 128000,
|
|
622
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/openai__gpt-6-luna/openai-gpt-6-luna-b5ddc970fc91fb39",
|
|
623
|
+
"sourceMaxTokens": 4000,
|
|
624
|
+
"sourceTargets": 3,
|
|
625
|
+
"promptTokens": 25444,
|
|
626
|
+
"completionTokens": 4546,
|
|
627
|
+
"estimateUsd": 0.07226099999999999,
|
|
628
|
+
"reservationUsd": 0.13425,
|
|
629
|
+
"capUsd": 0.5,
|
|
630
|
+
"estimatedMinutes": 56.0955,
|
|
631
|
+
"estimatedCalls": 360,
|
|
632
|
+
"spending": {
|
|
633
|
+
"version": "reserved-text-v2",
|
|
634
|
+
"limitUsd": 0.5,
|
|
635
|
+
"providerTag": "openai/flex",
|
|
636
|
+
"contextTokens": 1050000,
|
|
637
|
+
"inputUsdPerToken": 1.25e-7,
|
|
638
|
+
"outputUsdPerToken": 3.75e-7,
|
|
639
|
+
"model": "openai/gpt-6-luna",
|
|
640
|
+
"temperatureSupported": false,
|
|
641
|
+
"maxCompletionTokens": 128000,
|
|
642
|
+
"pricing": {
|
|
643
|
+
"completion": "0.00000025",
|
|
644
|
+
"discount": 0,
|
|
645
|
+
"input_cache_read": "0.000000005",
|
|
646
|
+
"input_cache_write": "0.0000000625",
|
|
647
|
+
"overrides": [
|
|
648
|
+
{
|
|
649
|
+
"completion": "0.000000375",
|
|
650
|
+
"input_cache_read": "0.00000001",
|
|
651
|
+
"input_cache_write": "0.000000125",
|
|
652
|
+
"min_prompt_tokens": 272000,
|
|
653
|
+
"prompt": "0.0000001"
|
|
654
|
+
}
|
|
655
|
+
],
|
|
656
|
+
"prompt": "0.00000005",
|
|
657
|
+
"web_search": "0.01"
|
|
658
|
+
}
|
|
659
|
+
},
|
|
660
|
+
"cliArgs": [
|
|
661
|
+
"read-the-room",
|
|
662
|
+
"openai/gpt-6-luna",
|
|
663
|
+
"--controls",
|
|
664
|
+
"--provider",
|
|
665
|
+
"openai",
|
|
666
|
+
"--split",
|
|
667
|
+
"evaluation",
|
|
668
|
+
"--limit",
|
|
669
|
+
"45",
|
|
670
|
+
"--budget",
|
|
671
|
+
"3",
|
|
672
|
+
"--seed",
|
|
673
|
+
"7",
|
|
674
|
+
"--max-model-calls",
|
|
675
|
+
"8",
|
|
676
|
+
"--max-tokens",
|
|
677
|
+
"8000",
|
|
678
|
+
"--agent-policy",
|
|
679
|
+
"transfer-check-v1",
|
|
680
|
+
"--max-usd",
|
|
681
|
+
"0.5",
|
|
682
|
+
"--output",
|
|
683
|
+
"results/final-8000/openai__gpt-6-luna"
|
|
684
|
+
]
|
|
685
|
+
},
|
|
686
|
+
{
|
|
687
|
+
"model": "anthropic/claude-haiku-5.5",
|
|
688
|
+
"provider": "anthropic",
|
|
689
|
+
"quantization": "unknown",
|
|
690
|
+
"catalogStatus": 0,
|
|
691
|
+
"source": "https://openrouter.ai/api/v1/models/anthropic/claude-haiku-5.5/endpoints",
|
|
692
|
+
"endpointPricing": {
|
|
693
|
+
"prompt": "0.0000001",
|
|
694
|
+
"completion": "0.0000005",
|
|
695
|
+
"web_search": "0.01",
|
|
696
|
+
"input_cache_read": "0.00000001",
|
|
697
|
+
"input_cache_write": "0.000000125",
|
|
698
|
+
"input_cache_write_1h": "0.0000002",
|
|
699
|
+
"discount": 0,
|
|
700
|
+
"overrides": [
|
|
701
|
+
{
|
|
702
|
+
"min_prompt_tokens": 100000,
|
|
703
|
+
"prompt": "0.0000005",
|
|
704
|
+
"completion": "0.0000025",
|
|
705
|
+
"input_cache_read": "0.00000005",
|
|
706
|
+
"input_cache_write": "0.000000625",
|
|
707
|
+
"input_cache_write_1h": "0.000001"
|
|
708
|
+
}
|
|
709
|
+
]
|
|
710
|
+
},
|
|
711
|
+
"temperatureSupported": false,
|
|
712
|
+
"maxCompletionTokens": 128000,
|
|
713
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/anthropic__claude-haiku-5.5/anthropic-claude-haiku-5-5-b5d2db2850447f5e",
|
|
714
|
+
"sourceMaxTokens": 4000,
|
|
715
|
+
"sourceTargets": 3,
|
|
716
|
+
"promptTokens": 37703,
|
|
717
|
+
"completionTokens": 4343,
|
|
718
|
+
"estimateUsd": 0.089127,
|
|
719
|
+
"reservationUsd": 1.02,
|
|
720
|
+
"capUsd": 1.5,
|
|
721
|
+
"estimatedMinutes": 11.076250000000002,
|
|
722
|
+
"estimatedCalls": 330,
|
|
723
|
+
"spending": {
|
|
724
|
+
"version": "reserved-text-v2",
|
|
725
|
+
"limitUsd": 1.5,
|
|
726
|
+
"providerTag": "anthropic",
|
|
727
|
+
"contextTokens": 1000000,
|
|
728
|
+
"inputUsdPerToken": 0.000001,
|
|
729
|
+
"outputUsdPerToken": 0.0000025,
|
|
730
|
+
"model": "anthropic/claude-haiku-5.5",
|
|
731
|
+
"temperatureSupported": false,
|
|
732
|
+
"maxCompletionTokens": 128000,
|
|
733
|
+
"pricing": {
|
|
734
|
+
"completion": "0.0000005",
|
|
735
|
+
"discount": 0,
|
|
736
|
+
"input_cache_read": "0.00000001",
|
|
737
|
+
"input_cache_write": "0.000000125",
|
|
738
|
+
"input_cache_write_1h": "0.0000002",
|
|
739
|
+
"overrides": [
|
|
740
|
+
{
|
|
741
|
+
"completion": "0.0000025",
|
|
742
|
+
"input_cache_read": "0.00000005",
|
|
743
|
+
"input_cache_write": "0.000000625",
|
|
744
|
+
"input_cache_write_1h": "0.000001",
|
|
745
|
+
"min_prompt_tokens": 100000,
|
|
746
|
+
"prompt": "0.0000005"
|
|
747
|
+
}
|
|
748
|
+
],
|
|
749
|
+
"prompt": "0.0000001",
|
|
750
|
+
"web_search": "0.01"
|
|
751
|
+
}
|
|
752
|
+
},
|
|
753
|
+
"cliArgs": [
|
|
754
|
+
"read-the-room",
|
|
755
|
+
"anthropic/claude-haiku-5.5",
|
|
756
|
+
"--controls",
|
|
757
|
+
"--provider",
|
|
758
|
+
"anthropic",
|
|
759
|
+
"--split",
|
|
760
|
+
"evaluation",
|
|
761
|
+
"--limit",
|
|
762
|
+
"45",
|
|
763
|
+
"--budget",
|
|
764
|
+
"3",
|
|
765
|
+
"--seed",
|
|
766
|
+
"7",
|
|
767
|
+
"--max-model-calls",
|
|
768
|
+
"8",
|
|
769
|
+
"--max-tokens",
|
|
770
|
+
"8000",
|
|
771
|
+
"--agent-policy",
|
|
772
|
+
"transfer-check-v1",
|
|
773
|
+
"--max-usd",
|
|
774
|
+
"1.5",
|
|
775
|
+
"--output",
|
|
776
|
+
"results/final-8000/anthropic__claude-haiku-5.5"
|
|
777
|
+
]
|
|
778
|
+
},
|
|
779
|
+
{
|
|
780
|
+
"model": "thinkingmachines/inkling-small",
|
|
781
|
+
"provider": "deepinfra/fp4",
|
|
782
|
+
"quantization": "fp4",
|
|
783
|
+
"catalogStatus": 0,
|
|
784
|
+
"source": "https://openrouter.ai/api/v1/models/thinkingmachines/inkling-small/endpoints",
|
|
785
|
+
"endpointPricing": {
|
|
786
|
+
"prompt": "0.00000045",
|
|
787
|
+
"completion": "0.0000012",
|
|
788
|
+
"input_cache_read": "0.0000001",
|
|
789
|
+
"discount": 0
|
|
790
|
+
},
|
|
791
|
+
"temperatureSupported": true,
|
|
792
|
+
"maxCompletionTokens": 262144,
|
|
793
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/thinkingmachines__inkling-small/thinkingmachines-inkling-small-d64c7fd43d84fe01",
|
|
794
|
+
"sourceMaxTokens": 4000,
|
|
795
|
+
"sourceTargets": 3,
|
|
796
|
+
"promptTokens": 25366,
|
|
797
|
+
"completionTokens": 6222,
|
|
798
|
+
"estimateUsd": 0.2832165,
|
|
799
|
+
"reservationUsd": 0.2455296,
|
|
800
|
+
"capUsd": 1.25,
|
|
801
|
+
"estimatedMinutes": 7.366499999999999,
|
|
802
|
+
"estimatedCalls": 360,
|
|
803
|
+
"spending": {
|
|
804
|
+
"version": "reserved-context-v1",
|
|
805
|
+
"limitUsd": 1.25,
|
|
806
|
+
"providerTag": "deepinfra/fp4",
|
|
807
|
+
"contextTokens": 524288,
|
|
808
|
+
"inputUsdPerToken": 4.5e-7,
|
|
809
|
+
"outputUsdPerToken": 0.0000012
|
|
810
|
+
},
|
|
811
|
+
"cliArgs": [
|
|
812
|
+
"read-the-room",
|
|
813
|
+
"thinkingmachines/inkling-small",
|
|
814
|
+
"--controls",
|
|
815
|
+
"--provider",
|
|
816
|
+
"deepinfra/fp4",
|
|
817
|
+
"--split",
|
|
818
|
+
"evaluation",
|
|
819
|
+
"--limit",
|
|
820
|
+
"45",
|
|
821
|
+
"--budget",
|
|
822
|
+
"3",
|
|
823
|
+
"--seed",
|
|
824
|
+
"7",
|
|
825
|
+
"--max-model-calls",
|
|
826
|
+
"8",
|
|
827
|
+
"--max-tokens",
|
|
828
|
+
"8000",
|
|
829
|
+
"--agent-policy",
|
|
830
|
+
"transfer-check-v1",
|
|
831
|
+
"--max-usd",
|
|
832
|
+
"1.25",
|
|
833
|
+
"--output",
|
|
834
|
+
"results/final-8000/thinkingmachines__inkling-small"
|
|
835
|
+
]
|
|
836
|
+
},
|
|
837
|
+
{
|
|
838
|
+
"model": "openai/gpt-oss-120b",
|
|
839
|
+
"provider": "primeintellect/fp4",
|
|
840
|
+
"quantization": "fp4",
|
|
841
|
+
"catalogStatus": 0,
|
|
842
|
+
"source": "https://openrouter.ai/api/v1/models/openai/gpt-oss-120b/endpoints",
|
|
843
|
+
"endpointPricing": {
|
|
844
|
+
"prompt": "0.000000035",
|
|
845
|
+
"completion": "0.000000075",
|
|
846
|
+
"input_cache_read": "0.000000031",
|
|
847
|
+
"discount": 0
|
|
848
|
+
},
|
|
849
|
+
"temperatureSupported": true,
|
|
850
|
+
"maxCompletionTokens": 65536,
|
|
851
|
+
"sourceDirectory": "results/development-preflight-2026-11-14/openai__gpt-oss-120b/openai-gpt-oss-120b-7eeb6507aa46c62f",
|
|
852
|
+
"sourceMaxTokens": 4000,
|
|
853
|
+
"sourceTargets": 3,
|
|
854
|
+
"promptTokens": 17433,
|
|
855
|
+
"completionTokens": 2446,
|
|
856
|
+
"estimateUsd": 0.011904075,
|
|
857
|
+
"reservationUsd": 0.00518752,
|
|
858
|
+
"capUsd": 0.25,
|
|
859
|
+
"estimatedMinutes": 7.274,
|
|
860
|
+
"estimatedCalls": 300,
|
|
861
|
+
"spending": {
|
|
862
|
+
"version": "reserved-context-v1",
|
|
863
|
+
"limitUsd": 0.25,
|
|
864
|
+
"providerTag": "primeintellect/fp4",
|
|
865
|
+
"contextTokens": 131072,
|
|
866
|
+
"inputUsdPerToken": 3.5e-8,
|
|
867
|
+
"outputUsdPerToken": 7.5e-8
|
|
868
|
+
},
|
|
869
|
+
"cliArgs": [
|
|
870
|
+
"read-the-room",
|
|
871
|
+
"openai/gpt-oss-120b",
|
|
872
|
+
"--controls",
|
|
873
|
+
"--provider",
|
|
874
|
+
"primeintellect/fp4",
|
|
875
|
+
"--split",
|
|
876
|
+
"evaluation",
|
|
877
|
+
"--limit",
|
|
878
|
+
"45",
|
|
879
|
+
"--budget",
|
|
880
|
+
"3",
|
|
881
|
+
"--seed",
|
|
882
|
+
"7",
|
|
883
|
+
"--max-model-calls",
|
|
884
|
+
"8",
|
|
885
|
+
"--max-tokens",
|
|
886
|
+
"8000",
|
|
887
|
+
"--agent-policy",
|
|
888
|
+
"transfer-check-v1",
|
|
889
|
+
"--max-usd",
|
|
890
|
+
"0.25",
|
|
891
|
+
"--output",
|
|
892
|
+
"results/final-8000/openai__gpt-oss-120b"
|
|
893
|
+
]
|
|
894
|
+
}
|
|
895
|
+
],
|
|
896
|
+
"totalEstimateUsd": 1.9729595133999995,
|
|
897
|
+
"totalCapUsd": 10,
|
|
898
|
+
"totalEstimatedMinutes": 353.9024833333333,
|
|
899
|
+
"totalEstimatedCalls": 3827
|
|
900
|
+
}
|