Benchmarks · agentic

EnterpriseOps-Gym-AA

EnterpriseOps-Gym-AA: multi-turn enterprise operations tasks over simulated business systems; success rate.

36 models · 38 results · data as of 2026-10-29 · artificialanalysis.ai ↗

Breakdown

darker is better within each column · click a column to sort

Model
Gemini 3.7 Flash medium	224,679,563	974,560	3,686,427	199,771,178
Gemini 3.5 Flash high	441,036,024	1,158,929	11,508,048	407,462,025
Grok 4.6 high	144,950,132	1,117,437	3,635,978	125,265,309
Opus 5 max	142,146,822	2,878,312	2,933,840	117,671,195
Qwen 3.8 2.4T-A95B	152,942,488	2,001,367	5,790,338	130,538,593
Muse Spark 1.2 xhigh	516,947,293	2,544,821	5,556,851	488,896,841
GPT-5.5 xhigh	124,443,990	910,393	4,184,480	113,275,357
Kimi K3 max	252,362,113	2,096,063	4,544,471	230,985,809
Qwen 3.7 Max	123,612,623	1,903,972	3,078,949	105,296,489
Sonnet 5 max	182,302,874	3,399,427	11,051,080	150,388,574
Qwen 3.8 27B xhigh	168,084,065	2,074,686	6,405,330	143,696,641
Opus 4.8 max	244,861,041	3,457,395	18,156,546	206,432,970
GPT-5.6 Sol max	118,768,640	781,848	2,926,551	107,793,624
GLM 5.2 max	187,741,219	2,211,534	6,749,954	165,615,855
Inkling Small	250,962,803	1,358,376	4,246,087	231,268,486
Gemini 3.5 Flash-Lite high	240,484,299	856,720	4,365,812	210,548,559
GPT-5.6 Luna max	121,824,316	824,287	3,843,153	105,826,854
Grok 4.5 high	332,994,695	1,765,407	2,600,528	316,292,141
Qwen 3.7 Plus	120,431,777	1,890,097	3,112,719	101,868,205
DeepSeek V4 Pro 0424 max	197,215,008	2,744,891	3,643,333	175,236,237
Kimi K2.7 Code	92,084,189	896,030	2,283,893	77,190,068
DeepSeek V4 Flash 0420 max	197,852,722	2,353,826	2,888,163	176,322,625
Kimi K2.6	116,009,498	849,328	3,472,328	99,552,324
GPT-5.6 Terra max	101,989,800	692,318	4,005,646	87,948,521
Inkling xhigh	286,018,135	1,487,726	2,796,062	263,862,859

All results

38 rows

#	Model	Effort	EnterpriseOps-Gym-AA
1	Gemini 3.7 Flash	medium	50.4%
2	Gemini 3.5 Flash	high	50.1%
3	Grok 4.6	high	48.3%
4	Opus 5	max	47.5%
5	Qwen 3.8 2.4T-A95B		47.4%
6	Muse Spark 1.2	xhigh	47.3%
7	GPT-5.5	xhigh	46.6%
8	Kimi K3	max	45.3%
9	Qwen 3.7 Max	max	45.0%
10	Sonnet 5	max	44.7%
11	Qwen 3.8 27B	xhigh	44.2%
12	GPT-5.5	medium	44.1%
13	Opus 4.8	max	44.0%
14	Sonnet 5	xhigh	43.9%
15	GPT-5.6 Sol	max	42.9%
16	GLM 5.2	max	42.7%
17	Inkling Small		42.7%
18	Gemini 3.5 Flash-Lite	high	42.3%
19	GPT-5.6 Luna	max	40.8%
20	Grok 4.5	high	40.8%
21	Qwen 3.7 Plus		40.6%
22	DeepSeek V4 Pro 0424	max	40.4%
23	Kimi K2.7 Code		40.2%
24	DeepSeek V4 Flash 0420	max	39.6%
25	Kimi K2.6		38.5%
26	GPT-5.6 Terra	max	38.5%
27	Inkling	xhigh	38.0%
28	GLM 5.3	max	36.4%
29	GPT-5.4 Mini	xhigh	34.6%
30	Mistral Medium 3.5	high	33.7%