Agents · Benchmark profile

EnterpriseOps-Gym-AA

Stateful multi-step enterprise workflow agent benchmark graded on final database state across business domains.

Ranking weightedCurrent / rollingActive25 models

Data verified 15 Jul 2026 · Methodology 2.3.0

Benchmark score on EnterpriseOps-Gym-AA

Leaderboard · 25 models

Claude Fable 5 leads at 51.1%, followed by Gemini 3.5 Flash (50.1%) and GPT-5.5 (46.6%).

Ranking · 0.9%AgentsCurrent

#1Closed

Claude Fable 5

Anthropic

claude-fable-5

51.1%

Overall #2 · Context 1M tokens

#2Closed

Gemini 3.5 Flash

Google

gemini-3.5-flash

50.1%

Overall #24 · Context 1.05M tokens

#3Closed

GPT-5.5

OpenAI

gpt-5.5

46.6%

Overall #13 · Context 1.05M tokens

ViewModelsColours

AnthropicGoogleOpenAIMoonshot AIAlibaba CloudZ.AIxAI

Lumina

EnterpriseOps-Gym-AA leaderboard

Accessible data for EnterpriseOps-Gym-AA leaderboard
Model	Rank	Model ID	Provider	Score	Unit
Claude Fable 5	#1	claude-fable-5	Anthropic	51.1191	percent
Gemini 3.5 Flash	#2	gemini-3-5-flash	Google	50.0746	percent
GPT-5.5	#3	gpt-5-5	OpenAI	46.6428	percent
Kimi K3	#4	kimi-k3	Moonshot AI	45.3	percent
Qwen3.7-Max	#5	qwen-3-7-max	Alibaba Cloud	45.0015	percent
Claude Sonnet 5	#6	claude-sonnet-5	Anthropic	44.6732	percent
Claude Opus 4.8	#7	claude-opus-4-8	Anthropic	43.957	percent
GLM-5.2	#8	glm-5-2	Z.AI	42.7335	percent
Gemini 3.1 Pro Preview	#9	gemini-3-1-pro-preview	Google	42.1665	percent
Grok 4.5	#10	grok-4-5	xAI	40.8236	percent
Qwen3.7-Plus	#11	qwen-3-7-plus	Alibaba Cloud	40.5551	percent
DeepSeek V4 Pro	#12	deepseek-v4-pro	DeepSeek	40.4357	percent
Kimi K2.7 Code	#13	kimi-k2-7-code	Moonshot AI	40.2268	percent
DeepSeek V4 Flash	#14	deepseek-v4-flash	DeepSeek	39.5703	percent
Kimi K2.6	#15	kimi-k2-6	Moonshot AI	38.5258	percent
GPT-5.4 mini	#16	gpt-5-4-mini	OpenAI	34.6165	percent
Mistral Medium 3.5	#17	mistral-medium-3-5	Mistral AI	33.7213	percent
MiniMax M3	#18	minimax-m3	MiniMax	32.1098	percent
GPT-5.4 nano	#19	gpt-5-4-nano	OpenAI	31.4831	percent
Qwen3.5 397B A17B	#20	qwen3-5-397b	Alibaba Cloud	30.8863	percent
Claude Haiku 4.5	#21	claude-haiku-4-5	Anthropic	29.3942	percent
Nemotron 3 Ultra	#22	nemotron-3-ultra	NVIDIA	28.8869	percent
Gemma 4 31B	#23	gemma-4-31b	Google	28.3497	percent
Gemini 3.1 Flash-Lite	#24	gemini-3-1-flash-lite	Google	28.0215	percent
Grok 4.3	#25	grok-4-3	xAI	26.2608	percent

Benchmark score table (25 models)

One best compatible score per canonical product · higher is better

EnterpriseOps-Gym-AA leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	Claude Fable 5 claude-fable-5	Anthropic	closed	Score input	51.1%
#2	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	Score input	50.1%
#3	GPT-5.5 gpt-5.5	OpenAI	closed	Score input	46.6%
#4	Kimi K3 kimi-k3	Moonshot AI	open	Score input	45.3%
#5	Qwen3.7-Max qwen3.7-max	Alibaba Cloud	closed	Score input	45.0%
#6	Claude Sonnet 5 claude-sonnet-5	Anthropic	closed	Score input	44.7%
#7	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	Score input	44.0%
#8	GLM-5.2 glm-5-2	Z.AI	open	Score input	42.7%
#9	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	Score input	42.2%
#10	Grok 4.5 grok-4.5	xAI	closed	Score input	40.8%
#11	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	closed	Score input	40.6%
#12	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	Score input	40.4%
#13	Kimi K2.7 Code kimi-k2.7-code	Moonshot AI	open	Score input	40.2%
#14	DeepSeek V4 Flash deepseek-v4-flash	DeepSeek	open	Score input	39.6%
#15	Kimi K2.6 kimi-k2.6	Moonshot AI	open	Score input	38.5%
#16	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	Score input	34.6%
#17	Mistral Medium 3.5 mistral-medium-3-5	Mistral AI	open	Score input	33.7%
#18	MiniMax M3 MiniMax-M3	MiniMax	open	Score input	32.1%
#19	GPT-5.4 nano gpt-5.4-nano	OpenAI	closed	Score input	31.5%
#20	Qwen3.5 397B A17B Qwen/Qwen3.5-397B-A17B	Alibaba Cloud	open	Score input	30.9%
#21	Claude Haiku 4.5 claude-haiku-4-5-20251001	Anthropic	closed	Score input	29.4%
#22	Nemotron 3 Ultra nemotron-3-ultra-550b	NVIDIA	open	Score input	28.9%
#23	Gemma 4 31B gemma-4-31b	Google	open	Score input	28.3%
#24	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	Score input	28.0%
#25	Grok 4.3 grok-4.3	xAI	closed	Score input	26.3%

The top of this snapshot is led by Claude Fable 5 at 51.1%; third place is 4.5 points behind. The top-10 spread is 10.3 points.

Configuration-preserving evidencePublished source rows (25)Exact source labels, configurations, versions and dated observations. Same-snapshot canonical/configuration projections are shown once.

Every distinct dated source-published row for EnterpriseOps-Gym-AA, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Kimi K3; Artificial Analysis EnterpriseOps-Gym-AA evaluation. source label without registered configuration ID	Kimi K3	45.3%	CurrentArtificial Analysis independent evaluation	Score input source-checked	Kimi K3 Intelligence, Performance & Price Analysis ↗Observed 2026-07-20 · checked 2026-07-21
Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) claude-fable-5-max	Claude Fable 5	51.1%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Gemini 3.5 Flash (high) gemini-3-5-flash-high	Gemini 3.5 Flash	50.1%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
GPT-5.5 (xhigh) gpt-5-5-xhigh	GPT-5.5	46.6%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.7 Max source label without registered configuration ID	Qwen3.7-Max	45.0%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Claude Sonnet 5 (Adaptive Reasoning, Max Effort) claude-sonnet-5-max	Claude Sonnet 5	44.7%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Claude Opus 4.8 (Adaptive Reasoning, Max Effort) claude-opus-4-8-max	Claude Opus 4.8	44.0%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
GLM-5.2 (max) glm-5-2-max	GLM-5.2	42.7%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Gemini 3.1 Pro Preview source label without registered configuration ID	Gemini 3.1 Pro Preview	42.2%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Grok 4.5 (high) source label without registered configuration ID	Grok 4.5	40.8%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.7 Plus source label without registered configuration ID	Qwen3.7-Plus	40.6%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
DeepSeek V4 Pro (Reasoning, Max Effort) deepseek-v4-pro-max	DeepSeek V4 Pro	40.4%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Kimi K2.7 Code source label without registered configuration ID	Kimi K2.7 Code	40.2%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
DeepSeek V4 Flash (Reasoning, Max Effort) deepseek-v4-flash-max	DeepSeek V4 Flash	39.6%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Kimi K2.6 source label without registered configuration ID	Kimi K2.6	38.5%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
GPT-5.4 mini (xhigh) gpt-5-4-mini-xhigh	GPT-5.4 mini	34.6%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Mistral Medium 3.5 source label without registered configuration ID	Mistral Medium 3.5	33.7%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
MiniMax-M3 source label without registered configuration ID	MiniMax M3	32.1%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
GPT-5.4 nano (xhigh) gpt-5-4-nano-xhigh	GPT-5.4 nano	31.5%	CurrentArtificial Analysis independent evaluation	Public reference · not scored source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Qwen3.5 397B A17B (Reasoning) source label without registered configuration ID	Qwen3.5 397B A17B	30.9%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Claude 4.5 Haiku (Reasoning) source label without registered configuration ID	Claude Haiku 4.5	29.4%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Nemotron 3 Ultra 550B A55B (Reasoning) source label without registered configuration ID	Nemotron 3 Ultra	28.9%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Gemma 4 31B (Reasoning) source label without registered configuration ID	Gemma 4 31B	28.3%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Gemini 3.1 Flash-Lite source label without registered configuration ID	Gemini 3.1 Flash-Lite	28.0%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15
Grok 4.3 (high) source label without registered configuration ID	Grok 4.3	26.3%	CurrentArtificial Analysis independent evaluation	Score input source-checked	EnterpriseOps-Gym-AA Leaderboard ↗Observed 2026-07-15 · checked 2026-07-15

Compare top models

Claude Fable 5 vs Gemini 3.5 Flash Gemini 3.5 Flash vs GPT-5.5 GPT-5.5 vs Kimi K3 Kimi K3 vs Qwen3.7-Max

About EnterpriseOps-Gym-AA

Definition and scoring

Organisation: Artificial Analysis / ServiceNow
Category: Agents
Version: Not evaluated
Direction: higher is better
Ranking use: Weighted 0.9%
Contamination risk: Unknown

Stateful multi-step enterprise workflow agent benchmark graded on final database state across business domains. Every genuine source row stays tied to its exact model label, configuration, benchmark version and evaluation system. The summary chart shows one best compatible score per canonical product; Score 2.0 admits only explicitly mapped, frozen protocols and keeps incompatible configurations separate.

Open benchmark source ↗

FAQ

Quick answers

What does EnterpriseOps-Gym-AA measure?

Stateful multi-step enterprise workflow agent benchmark graded on final database state across business domains.

Which model scores highest on EnterpriseOps-Gym-AA?

Claude Fable 5 by Anthropic currently leads with 51.1%.

How many models are evaluated on EnterpriseOps-Gym-AA?

25 canonical model products have a qualifying score, backed by 25 distinct dated source rows.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 0.9% of its capability category. Claude Fable 5 is overall #2.

Agents · Benchmark profile

EnterpriseOps-Gym-AA

Stateful multi-step enterprise workflow agent benchmark graded on final database state across business domains.

Ranking weightedCurrent / rollingActive25 models

Data verified 15 Jul 2026 · Methodology 2.3.0

Benchmark score on EnterpriseOps-Gym-AA