Agentic · tools

AA EnterpriseOps-Gym leaderboard

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

As of 2026-10-07, the highest measured score on AA EnterpriseOps-Gym is 51.1% by Claude Fable 5. 196 more models have estimated scores, calibrated from the benchmarks they were measured on.

Measured scores: benchlm.ai.

#	Model	Score	Source
1	Gemini 3.7 Flash	57.6%	estimated ± 1.8 pp, low confidence
2	Claude Fable 5.1	56.3%	estimated ± 1.8 pp, low confidence
3	Claude Sonnet 5.5	56.3%	estimated ± 1.8 pp, low confidence
4	Claude Opus 5.5	55.6%	estimated ± 1.8 pp, low confidence
5	Claude Opus 5	54.3%	estimated ± 1.8 pp, low confidence
6	GLM-5.2	53.8%	estimated ± 3.6 pp, low confidence
7	GPT-5.4	53.5%	estimated ± 3.6 pp, low confidence
8	GPT-6 Astra	53.0%	estimated ± 1.8 pp, low confidence
9	GLM-5-Turbo	52.8%	estimated ± 3.6 pp, medium confidence
10	GLM-5V-Turbo	52.8%	estimated ± 3.6 pp, medium confidence
11	Step 3.7 Flash	52.8%	estimated ± 3.6 pp, medium confidence
12	GPT-5.5	52.4%	estimated ± 1.8 pp, low confidence
13	GPT-6.1 Sol	52.4%	estimated ± 1.8 pp, low confidence
14	GLM-5	52.2%	estimated ± 3.6 pp, medium confidence
15	Muse Spark 1.3	51.9%	estimated ± 4.4 pp, medium confidence
16	GLM-5.1	51.4%	estimated ± 3.6 pp, medium confidence
17	Grok 4.3	51.4%	estimated ± 3.6 pp, medium confidence
18	Qwen3.6 Plus	51.4%	estimated ± 3.6 pp, medium confidence
19	Claude Fable 5	51.1%	measured
20	GPT-5.6 Sol	51.0%	estimated ± 1.8 pp, medium confidence
21	GPT-6 Sol	50.4%	estimated ± 4.4 pp, medium confidence
22	Gemini 3.5 Flash	50.1%	measured
23	Claude Mythos 5	49.9%	estimated ± 5.9 pp, low confidence
24	DeepSeek V4 Pro 0813	49.6%	measured
25	DeepSeek V4 Flash 0731	49.4%	estimated ± 5.1 pp, low confidence
26	Qwen3.8 Max	49.1%	estimated ± 5.1 pp, low confidence
27	Grok 4.6	48.3%	measured
28	Hy4 preview	48.3%	estimated ± 5.1 pp, low confidence
29	GLM-4.7	48.1%	estimated ± 3.6 pp, medium confidence
30	Kimi K2.6	48.1%	estimated ± 3.6 pp, medium confidence
31	Kimi K2.5	48.1%	estimated ± 3.6 pp, medium confidence
32	Kimi K2.5 (Reasoning)	48.1%	estimated ± 3.6 pp, medium confidence
33	Qwen 3.6 Max (preview)	48.1%	estimated ± 3.6 pp, medium confidence
34	GPT-6 Luna	47.9%	estimated ± 4.4 pp, medium confidence
35	Gemini 3.6 Flash	47.9%	estimated ± 5.9 pp, low confidence
36	Gemini 3.1 Pro	47.6%	estimated ± 3.6 pp, medium confidence
37	Holo3-35B-A3B	47.4%	estimated ± 5.9 pp, low confidence
38	Qwen3.8 Max Preview	47.3%	estimated ± 6.7 pp, low confidence
39	Step 5 Preview	47.2%	measured
40	Qwen3.6-35B-A3B	47.1%	estimated ± 3.6 pp, medium confidence
41	Qwen3.8-Flash-Next	47.0%	estimated ± 6.7 pp, low confidence
42	Gemini 4 Argon	46.9%	estimated ± 4.4 pp, high confidence
43	MiMo-V2-Pro	46.6%	estimated ± 3.6 pp, medium confidence
44	Claude Sonnet 5	46.1%	estimated ± 5.9 pp, low confidence
45	Gemini 3.8 Flash	46.1%	estimated ± 4.4 pp, high confidence
46	Qwen3.7 Max	46.0%	estimated ± 3.6 pp, medium confidence
47	Claude Haiku 5.5	45.8%	estimated ± 4.4 pp, high confidence
48	Muse Spark 1.1	45.7%	estimated ± 5.9 pp, low confidence
49	Claude Opus 4.8	45.5%	estimated ± 3.6 pp, medium confidence
50	Kimi K3	45.3%	measured
51	MiMo-V2.5-Pro	45.1%	estimated ± 3.6 pp, medium confidence
52	Mistral Medium 3.5 128B	45.1%	estimated ± 3.6 pp, medium confidence
53	Qwen3.6-27B	45.1%	estimated ± 3.6 pp, medium confidence
54	Grok 4.7	45.0%	estimated ± 4.4 pp, high confidence
55	Muse Spark 1.2	44.8%	estimated ± 6.7 pp, low confidence
56	Qwen3.5-27B	44.6%	estimated ± 3.6 pp, medium confidence
57	GPT-5.6 Luna	44.5%	estimated ± 6.7 pp, low confidence
58	GPT-5.5 Pro	44.3%	estimated ± 6.9 pp, low confidence
59	Qwen3.8-27B	44.2%	measured
60	MiMo-V2.6-Pro	44.2%	estimated ± 4.4 pp, high confidence
61	Qwen3.5-122B-A10B	44.1%	estimated ± 3.6 pp, medium confidence
62	GPT-5.4 Pro	44.1%	estimated ± 6.9 pp, low confidence
63	Holo3-122B-A10B	43.9%	estimated ± 5.9 pp, low confidence
64	GPT-5.4 mini	43.7%	estimated ± 3.6 pp, medium confidence
65	Grok 4.1 Fast (Reasoning)	43.6%	estimated ± 3.6 pp, medium confidence
66	Mistral Large 4	43.6%	estimated ± 4.4 pp, high confidence
67	Ornith-1.5-397B	43.2%	estimated ± 6.9 pp, low confidence
68	Qwen3.7 Plus	43.0%	estimated ± 3.6 pp, medium confidence
69	Grok 4.5	42.7%	estimated ± 6.7 pp, low confidence
70	UI-Mate-27B	42.2%	estimated ± 5.9 pp, low confidence
71	GPT-5.4 nano	42.2%	estimated ± 3.6 pp, medium confidence
72	dots3-note Preview	42.1%	estimated ± 6.9 pp, low confidence
73	Claude Opus 4.6 (Adaptive)	41.5%	estimated ± 3.6 pp, medium confidence
74	GPT-5.2-Codex	41.5%	estimated ± 3.6 pp, medium confidence
75	Muse Spark	40.4%	estimated ± 3.6 pp, medium confidence
76	Beam	40.1%	estimated ± 6.9 pp, low confidence
77	MiMo-V2-Omni	39.9%	estimated ± 3.6 pp, medium confidence
78	Gemini 3.5 Flash-Lite	39.6%	estimated ± 5.9 pp, low confidence
79	Agents-A1	39.5%	estimated ± 6.9 pp, low confidence
80	Kimi K2.7 Code	38.0%	estimated ± 3.6 pp, medium confidence
81	Trinity-Large-Preview	38.0%	estimated ± 3.6 pp, medium confidence
82	Trinity-Large-Thinking	38.0%	estimated ± 3.6 pp, medium confidence
83	Inkling	38.0%	measured
84	Hy3 Preview	37.5%	estimated ± 6.7 pp, low confidence
85	DeepSeek V4.1 Flash	37.5%	estimated ± 4.4 pp, high confidence
86	Ling 3.0 Flash	37.1%	estimated ± 5.3 pp, medium confidence
87	Claude Opus 4.5 Thinking	37.0%	estimated ± 3.6 pp, medium confidence
88	Apodex 1.1	37.0%	estimated ± 6.7 pp, low confidence
89	Apodex 1.1 Mini	37.0%	estimated ± 6.7 pp, low confidence
90	Ornith-1.5-35B-A3B	36.8%	estimated ± 6.9 pp, low confidence
91	Quasar 438B	36.6%	estimated ± 6.7 pp, low confidence
92	Qwen3.5-35B-A3B	36.5%	estimated ± 3.6 pp, medium confidence
93	GLM-5.3	36.4%	measured
94	Ling 3.0 Flash VL	36.1%	estimated ± 6.7 pp, low confidence
95	Claude Opus 4.7 (Adaptive)	35.5%	estimated ± 3.6 pp, medium confidence
96	Inkling-Small	35.2%	estimated ± 6.7 pp, low confidence
97	Solar Pro 4	34.9%	estimated ± 6.7 pp, low confidence
98	Muse Glimmer 30B	34.7%	measured
99	LFM2.5-8B-A1B	34.6%	estimated ± 3.6 pp, medium confidence
100	Hy3	34.1%	estimated ± 6.7 pp, low confidence
101	UI-Mate-9B	33.4%	estimated ± 5.9 pp, low confidence
102	GLM-5.3-Flash	33.2%	measured
103	Gemini 3 Pro	33.1%	estimated ± 3.6 pp, medium confidence
104	A.X K2	33.0%	estimated ± 6.7 pp, low confidence
105	Ling 3.0 Flash FP8	32.9%	estimated ± 6.7 pp, low confidence
106	Ornith-1.5-9B	32.7%	estimated ± 6.9 pp, low confidence
107	MiniCPM5-2B	32.5%	estimated ± 6.7 pp, low confidence
108	Nemotron 3.5 Lightning 30B A3B NVFP4	32.5%	estimated ± 6.7 pp, low confidence
109	Celeris-1	32.5%	estimated ± 6.7 pp, low confidence
110	GPT-4o mini	32.5%	estimated ± 6.7 pp, low confidence
111	Granite 4.2 30B	32.5%	estimated ± 6.7 pp, low confidence
112	Granite 4.2 3B	32.5%	estimated ± 6.7 pp, low confidence
113	Granite 4.2 8B	32.5%	estimated ± 6.7 pp, low confidence
114	LFM2.5-2.6B	32.5%	estimated ± 6.7 pp, low confidence
115	Ling 3.0 Tiny	32.5%	estimated ± 6.7 pp, low confidence
116	Mercury 2.5	32.5%	estimated ± 6.7 pp, low confidence
117	GPT-5 (medium)	32.1%	estimated ± 3.6 pp, medium confidence
118	MiniMax M3	32.1%	measured
119	MiMo-V2.6-Flash	31.9%	estimated ± 5.1 pp, low confidence
120	Claude Opus 4.5	31.8%	estimated ± 3.6 pp, medium confidence
121	GPT-5.6 Terra	31.8%	estimated ± 3.6 pp, medium confidence
122	Solar Pro 3	31.8%	estimated ± 3.6 pp, medium confidence
123	Ling 3.1 Flash	31.7%	estimated ± 5.1 pp, low confidence
124	GPT-5.3 Codex	31.3%	estimated ± 3.6 pp, medium confidence
125	Ling 2.6 Flash	31.3%	estimated ± 3.6 pp, medium confidence
126	Atria Dawn Preview	30.0%	estimated ± 5.1 pp, low confidence
127	Claude Sonnet 4.5	29.9%	estimated ± 5.9 pp, low confidence
128	LongCat-Flash-Lite-Sparse	29.8%	estimated ± 6.9 pp, low confidence
129	Command A+	29.8%	estimated ± 3.6 pp, medium confidence
130	Claude Opus 4.6	29.5%	estimated ± 3.6 pp, medium confidence
131	GPT-5.2	29.5%	estimated ± 3.6 pp, medium confidence
132	GPT-5 (high)	29.5%	estimated ± 3.6 pp, medium confidence
133	MiniMax M2.7	29.5%	estimated ± 3.6 pp, medium confidence
134	Nemotron 3 Ultra	28.9%	measured
135	MiMo-V2-Flash	28.1%	estimated ± 3.6 pp, medium confidence
136	Qwen3.5 397B	28.1%	estimated ± 3.6 pp, medium confidence
137	Qwen3.5 397B (Reasoning)	28.1%	estimated ± 3.6 pp, medium confidence
138	GPT-5.1-Codex	26.8%	estimated ± 3.6 pp, low confidence
139	GPT-5.1-Codex-Max	26.8%	estimated ± 3.6 pp, low confidence
140	GPT-5.1	25.2%	estimated ± 3.6 pp, low confidence
141	o3	23.6%	estimated ± 3.6 pp, low confidence
142	LLaDA2.2-flash	23.1%	estimated ± 3.6 pp, low confidence
143	Ternary Bonsai 2 27B	22.9%	estimated ± 3.6 pp, low confidence
144	Claude Sonnet 4.6	22.0%	estimated ± 3.6 pp, low confidence
145	DeepSeek V3.2	21.2%	estimated ± 3.6 pp, low confidence
146	Agents-A1-4B	20.3%	estimated ± 3.6 pp, low confidence
147	GLM-4.6	18.8%	estimated ± 3.6 pp, low confidence
148	Grok Code Fast 1	17.4%	estimated ± 3.6 pp, low confidence
149	Grok 4	16.5%	estimated ± 3.6 pp, low confidence
150	K-Exaone	15.9%	estimated ± 3.6 pp, low confidence
151	Qwen3 Max	15.9%	estimated ± 3.6 pp, low confidence
152	Claude Opus 4.7	15.6%	estimated ± 3.6 pp, low confidence
153	Claude 4.1 Opus Thinking	13.1%	estimated ± 3.6 pp, low confidence
154	Nemotron 3 Super 100B	10.1%	estimated ± 3.6 pp, low confidence
155	GPT-OSS 120B	8.7%	estimated ± 3.6 pp, low confidence
156	Grok 4 Fast (Reasoning)	8.7%	estimated ± 3.6 pp, low confidence
157	Grok 4.1 Fast	7.5%	estimated ± 3.6 pp, low confidence
158	o1	6.9%	estimated ± 3.6 pp, low confidence
159	Kimi K2	6.1%	estimated ± 3.6 pp, low confidence
160	GPT-OSS 20B	5.7%	estimated ± 3.6 pp, low confidence
161	Gemma 4 31B	5.6%	estimated ± 3.6 pp, low confidence
162	LLaDA2.2-mini	4.7%	estimated ± 3.6 pp, low confidence
163	Gemini 2.5 Pro	3.6%	estimated ± 3.6 pp, low confidence
164	GPT-4.1 mini	3.3%	estimated ± 3.6 pp, low confidence
165	Claude 4 Sonnet	3.2%	estimated ± 3.6 pp, low confidence
166	DeepSeek V3 0324	2.2%	estimated ± 3.6 pp, low confidence
167	GPT-4.1	2.2%	estimated ± 3.6 pp, low confidence
168	Sarvam 105B	2.2%	estimated ± 3.6 pp, low confidence
169	GLM-4.5-Air	2.2%	estimated ± 3.6 pp, low confidence
170	Nemotron 3 Nano Omni 30B A3B	2.0%	estimated ± 3.6 pp, low confidence
171	Gemma 4 26B A4B	1.8%	estimated ± 3.6 pp, low confidence
172	Gemini 3 Flash	1.8%	estimated ± 3.6 pp, low confidence
173	Mistral Small 4	1.6%	estimated ± 3.6 pp, low confidence
174	Mistral Small 4 (Reasoning)	1.6%	estimated ± 3.6 pp, low confidence
175	Nemotron 3 Nano 30B	1.6%	estimated ± 3.6 pp, low confidence
176	DeepSeek V3.1 (Reasoning)	1.4%	estimated ± 3.6 pp, low confidence
177	North Mini Code	1.4%	estimated ± 3.6 pp, low confidence
178	DeepSeek-R1	1.4%	estimated ± 3.6 pp, low confidence
179	Gemma 4 12B	1.4%	estimated ± 3.6 pp, low confidence
180	DeepSeek V3.1	1.3%	estimated ± 3.6 pp, low confidence
181	Sarvam 30B	1.3%	estimated ± 3.6 pp, low confidence
182	Solar Pro 2	1.3%	estimated ± 3.6 pp, low confidence
183	Mistral Large 2	1.2%	estimated ± 3.6 pp, low confidence
184	o3-mini	1.2%	estimated ± 3.6 pp, low confidence
185	Ultravox v0.6 Llama 3.3 70B	1.2%	estimated ± 3.6 pp, low confidence
186	GPT-4o	1.2%	estimated ± 3.6 pp, low confidence
187	Mistral Large 3	1.2%	estimated ± 3.6 pp, low confidence
188	Mistral Medium 3	1.2%	estimated ± 3.6 pp, low confidence
189	DeepSeek V3	1.2%	estimated ± 3.6 pp, low confidence
190	Qwen3-Omni-30B-A3B-Thinking	1.2%	estimated ± 3.6 pp, low confidence
191	Claude 3 Haiku	1.2%	estimated ± 3.6 pp, low confidence
192	Gemma 4 E2B	1.2%	estimated ± 3.6 pp, low confidence
193	Gemma 4 E4B	1.2%	estimated ± 3.6 pp, low confidence
194	Exaone 4.0 1.2B	1.2%	estimated ± 3.6 pp, low confidence
195	Granite-4.0-H-1B	1.2%	estimated ± 3.6 pp, low confidence
196	Llama 3.1 405B	1.2%	estimated ± 3.6 pp, low confidence
197	Llama 4 Maverick	1.2%	estimated ± 3.6 pp, low confidence
198	GPT-4.1 nano	1.2%	estimated ± 3.6 pp, low confidence
199	Qwen3-Omni-30B-A3B-Instruct	1.2%	estimated ± 3.6 pp, low confidence
200	Llama 4 Scout	1.2%	estimated ± 3.6 pp, low confidence
201	Gemini 2.5 Flash	1.2%	estimated ± 3.6 pp, low confidence
202	Granite-4.0-H-350M	1.2%	estimated ± 3.6 pp, low confidence
203	Nova Pro	1.2%	estimated ± 3.6 pp, low confidence
204	Granite-4.0-350M	1.2%	estimated ± 3.6 pp, low confidence
205	Nemotron Ultra 253B	1.2%	estimated ± 3.6 pp, low confidence
206	Gemma 3 27B	1.2%	estimated ± 3.6 pp, low confidence
207	Exaone 4.0 32B	1.2%	estimated ± 3.6 pp, low confidence
208	LFM2.5-VL-1.6B-Extract	1.2%	estimated ± 3.6 pp, low confidence
209	Phi-4	1.2%	estimated ± 3.6 pp, low confidence

AA EnterpriseOps-Gym leaderboard

All leaderboards

Agentic · terminal

Agentic · tools

Coding

Math

Knowledge & reasoning

Instruction following

Multilingual

Vision & documents

Long context

General