Agents · Benchmark profile

APEX-Agents-AA

Long-horizon professional multi-application agent benchmark implemented independently by Artificial Analysis.

Ranking weightedCurrent / rollingActive24 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on APEX-Agents-AA

Leaderboard · 24 models

Gemini 3.5 Flash leads at 47.1%, followed by GPT-5.5 (37.7%) and Kimi K3 (37.6%).

Ranking · 1.0%AgentsCurrent

#1Closed

Gemini 3.5 Flash

Google

gemini-3.5-flash

47.1%

Overall #17 · Context 1.05M tokens

#2Closed

GPT-5.5

OpenAI

gpt-5.5

37.7%

Overall #8 · Context 1.05M tokens

#3Closed

Kimi K3

Moonshot AI

kimi-k3

37.6%

Overall #3 · Context 1M tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

GoogleOpenAIMoonshot AIZ.AIAnthropicTencentDeepSeekAlibaba CloudxAI

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
Gemini 3.5 Flash	#1	Google	47.1%
GPT-5.5	#2	OpenAI	37.7%
Kimi K3	#3	Moonshot AI	37.6%
GLM-5.2	#4	Z.AI	33.7%
GPT-5.4	#5	OpenAI	33.3%
Claude Opus 4.6	#6	Anthropic	33.0%
Gemini 3.1 Pro Preview	#7	Google	32.0%
Kimi K2.6	#8	Moonshot AI	28.5%
GPT-5.4 mini	#9	OpenAI	28.2%
Claude Sonnet 4.6	#10	Anthropic	28.0%
Gemini 3 Flash	#11	Google	27.7%
Hy3	#12	Tencent	25.6%
GPT-5.4 nano	#13	OpenAI	24.9%
DeepSeek V4 Pro	#14	DeepSeek	24.3%
Qwen3.7-Plus	#15	Alibaba Cloud	22.4%
Grok 4.3	#16	xAI	17.0%
Qwen3.5 397B	#17	Alibaba Cloud	15.3%
DeepSeek V3.2	#18	DeepSeek	14.5%
GLM-5	#19	Z.AI	14.5%
Gemini 3.1 Flash-Lite	#20	Google	12.2%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
Gemini 3.5 Flash	#1	Google	47.1%
GPT-5.5	#2	OpenAI	37.7%
Kimi K3	#3	Moonshot AI	37.6%
GLM-5.2	#4	Z.AI	33.7%
GPT-5.4	#5	OpenAI	33.3%
Claude Opus 4.6	#6	Anthropic	33.0%
Gemini 3.1 Pro Preview	#7	Google	32.0%
Kimi K2.6	#8	Moonshot AI	28.5%
GPT-5.4 mini	#9	OpenAI	28.2%
Claude Sonnet 4.6	#10	Anthropic	28.0%
Gemini 3 Flash	#11	Google	27.7%
Hy3	#12	Tencent	25.6%
GPT-5.4 nano	#13	OpenAI	24.9%
DeepSeek V4 Pro	#14	DeepSeek	24.3%
Qwen3.7-Plus	#15	Alibaba Cloud	22.4%
Grok 4.3	#16	xAI	17.0%
Qwen3.5 397B	#17	Alibaba Cloud	15.3%
DeepSeek V3.2	#18	DeepSeek	14.5%
GLM-5	#19	Z.AI	14.5%
Gemini 3.1 Flash-Lite	#20	Google	12.2%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Moonshot AI	2	33.0%
OpenAI	4	31.0%
Anthropic	2	30.5%
Google	4	29.7%
Tencent	1	25.6%
Z.AI	2	24.1%
DeepSeek	2	19.4%
Alibaba Cloud	2	18.9%
xAI	1	17.0%

Benchmark score table (24 models)

One best score per model · higher is better

APEX-Agents-AA leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	47.1%
#2	GPT-5.5 gpt-5.5	OpenAI	closed	37.7%
#3	Kimi K3 kimi-k3	Moonshot AI	closed	37.6%
#4	GLM-5.2 glm-5-2	Z.AI	open	33.7%
#5	GPT-5.4 gpt-5.4	OpenAI	closed	33.3%
#6	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	33.0%
#7	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	32.0%
#8	Kimi K2.6 kimi-k2.6	Moonshot AI	open	28.5%
#9	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	28.2%
#10	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	28.0%
#11	Gemini 3 Flash gemini-3-flash	Google	closed	27.7%
#12	Hy3 hy3	Tencent	open	25.6%
#13	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	24.9%
#14	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	24.3%
#15	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	22.4%
#16	Grok 4.3 grok-4.3	xAI	closed	17.0%
#17	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	15.3%
#18	DeepSeek V3.2 deepseek-v3.2	DeepSeek	open	14.5%
#19	GLM-5 glm-5	Z.AI	open	14.5%
#20	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	12.2%
#21	Kimi K2.5 kimi-k2.5	Moonshot AI	—	11.5%
#22	MiniMax M2.7 minimax-m2-7	MiniMax	open	10.6%
#23	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	2.4%
#24	Nemotron 3 Super nemotron-3-super-120b	NVIDIA	open	1.8%

The top of this snapshot is led by Gemini 3.5 Flash at 47.1%; third place is 9.5 points behind. The top-10 spread is 19.0 points.

Compare top models

Gemini 3.5 Flash vs GPT-5.5 GPT-5.5 vs Kimi K3 Kimi K3 vs GLM-5.2 GLM-5.2 vs GPT-5.4

About APEX-Agents-AA

Definition and scoring

Organisation: Artificial Analysis / Mercor
Category: Agents
Version: Not evaluated
Direction: higher is better
Ranking use: Weighted 1.0%
Contamination risk: Unknown

Long-horizon professional multi-application agent benchmark implemented independently by Artificial Analysis. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.

Open benchmark source ↗

FAQ

Quick answers

What does APEX-Agents-AA measure?

Long-horizon professional multi-application agent benchmark implemented independently by Artificial Analysis.

Which model scores highest on APEX-Agents-AA?

Gemini 3.5 Flash by Google currently leads with 47.1%.

How many models are evaluated on APEX-Agents-AA?

24 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 1.0% of its capability category. Gemini 3.5 Flash is overall #17.

Terminal-Bench

2.1 · 80 results · Weighted 1.6%

Berkeley Function-Calling Leaderboard

V3 · 2 results · Not Weighted

τ-bench

Rolling · 37 results · Weighted 1.6%

GAIA

Rolling · 13 results · Weighted 1.6%

BrowseComp

Rolling · 24 results · Weighted 1.6%

OSWorld

Rolling · 11 results · Weighted 1.6%

Agents · Benchmark profile

APEX-Agents-AA

Long-horizon professional multi-application agent benchmark implemented independently by Artificial Analysis.

Ranking weightedCurrent / rollingActive24 models

Data verified 15 July 2026 · Methodology 1.2.0

Benchmark score on APEX-Agents-AA

Leaderboard · 24 models

Gemini 3.5 Flash leads at 47.1%, followed by GPT-5.5 (37.7%) and Kimi K3 (37.6%).

Ranking · 1.0%AgentsCurrent

#1Closed

Gemini 3.5 Flash

Google

gemini-3.5-flash

47.1%

Overall #17 · Context 1.05M tokens

#2Closed

GPT-5.5

OpenAI

gpt-5.5

37.7%

Overall #8 · Context 1.05M tokens

#3Closed

Kimi K3

Moonshot AI

kimi-k3

37.6%

Overall #3 · Context 1M tokens

Visual analysis

See the field, not just the table.

Three views of the same sourced leaderboard: model placement, the score curve, and descriptive provider averages. These visuals never alter ranking calculations.

ModelsColours

GoogleOpenAIMoonshot AIZ.AIAnthropicTencentDeepSeekAlibaba CloudxAI

Leaderboard bars

20 ranked models · higher is better

View accessible chart data

Accessible data for Leaderboard bars
Model	Rank	Provider	Score
Gemini 3.5 Flash	#1	Google	47.1%
GPT-5.5	#2	OpenAI	37.7%
Kimi K3	#3	Moonshot AI	37.6%
GLM-5.2	#4	Z.AI	33.7%
GPT-5.4	#5	OpenAI	33.3%
Claude Opus 4.6	#6	Anthropic	33.0%
Gemini 3.1 Pro Preview	#7	Google	32.0%
Kimi K2.6	#8	Moonshot AI	28.5%
GPT-5.4 mini	#9	OpenAI	28.2%
Claude Sonnet 4.6	#10	Anthropic	28.0%
Gemini 3 Flash	#11	Google	27.7%
Hy3	#12	Tencent	25.6%
GPT-5.4 nano	#13	OpenAI	24.9%
DeepSeek V4 Pro	#14	DeepSeek	24.3%
Qwen3.7-Plus	#15	Alibaba Cloud	22.4%
Grok 4.3	#16	xAI	17.0%
Qwen3.5 397B	#17	Alibaba Cloud	15.3%
DeepSeek V3.2	#18	DeepSeek	14.5%
GLM-5	#19	Z.AI	14.5%
Gemini 3.1 Flash-Lite	#20	Google	12.2%

Score curve

Shows how quickly scores separate down the field

View accessible chart data

Accessible data for Score curve
Model	Rank	Provider	Score
Gemini 3.5 Flash	#1	Google	47.1%
GPT-5.5	#2	OpenAI	37.7%
Kimi K3	#3	Moonshot AI	37.6%
GLM-5.2	#4	Z.AI	33.7%
GPT-5.4	#5	OpenAI	33.3%
Claude Opus 4.6	#6	Anthropic	33.0%
Gemini 3.1 Pro Preview	#7	Google	32.0%
Kimi K2.6	#8	Moonshot AI	28.5%
GPT-5.4 mini	#9	OpenAI	28.2%
Claude Sonnet 4.6	#10	Anthropic	28.0%
Gemini 3 Flash	#11	Google	27.7%
Hy3	#12	Tencent	25.6%
GPT-5.4 nano	#13	OpenAI	24.9%
DeepSeek V4 Pro	#14	DeepSeek	24.3%
Qwen3.7-Plus	#15	Alibaba Cloud	22.4%
Grok 4.3	#16	xAI	17.0%
Qwen3.5 397B	#17	Alibaba Cloud	15.3%
DeepSeek V3.2	#18	DeepSeek	14.5%
GLM-5	#19	Z.AI	14.5%
Gemini 3.1 Flash-Lite	#20	Google	12.2%

Provider profile

Average within the visible cohort · descriptive only

View accessible chart data

Accessible data for Provider profile
Provider	Models	Average score
Moonshot AI	2	33.0%
OpenAI	4	31.0%
Anthropic	2	30.5%
Google	4	29.7%
Tencent	1	25.6%
Z.AI	2	24.1%
DeepSeek	2	19.4%
Alibaba Cloud	2	18.9%
xAI	1	17.0%

Benchmark score table (24 models)

One best score per model · higher is better

APEX-Agents-AA leaderboard ranked by score
Rank	Model	Provider	License	Score
#1	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	47.1%
#2	GPT-5.5 gpt-5.5	OpenAI	closed	37.7%
#3	Kimi K3 kimi-k3	Moonshot AI	closed	37.6%
#4	GLM-5.2 glm-5-2	Z.AI	open	33.7%
#5	GPT-5.4 gpt-5.4	OpenAI	closed	33.3%
#6	Claude Opus 4.6 claude-opus-4-6	Anthropic	closed	33.0%
#7	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	32.0%
#8	Kimi K2.6 kimi-k2.6	Moonshot AI	open	28.5%
#9	GPT-5.4 mini gpt-5.4-mini	OpenAI	closed	28.2%
#10	Claude Sonnet 4.6 claude-sonnet-4-6	Anthropic	closed	28.0%
#11	Gemini 3 Flash gemini-3-flash	Google	closed	27.7%
#12	Hy3 hy3	Tencent	open	25.6%
#13	GPT-5.4 nano gpt.5.4.nano	OpenAI	closed	24.9%
#14	DeepSeek V4 Pro deepseek-v4-pro	DeepSeek	open	24.3%
#15	Qwen3.7-Plus qwen3.7-plus	Alibaba Cloud	—	22.4%
#16	Grok 4.3 grok-4.3	xAI	closed	17.0%
#17	Qwen3.5 397B qwen3.5-397b-a17b	Alibaba Cloud	open	15.3%
#18	DeepSeek V3.2 deepseek-v3.2	DeepSeek	open	14.5%
#19	GLM-5 glm-5	Z.AI	open	14.5%
#20	Gemini 3.1 Flash-Lite gemini-3.1-flash-lite	Google	closed	12.2%
#21	Kimi K2.5 kimi-k2.5	Moonshot AI	—	11.5%
#22	MiniMax M2.7 minimax-m2-7	MiniMax	open	10.6%
#23	MiMo-V2.5-Pro mimo-v2-5-pro	Xiaomi	closed	2.4%
#24	Nemotron 3 Super nemotron-3-super-120b	NVIDIA	open	1.8%

The top of this snapshot is led by Gemini 3.5 Flash at 47.1%; third place is 9.5 points behind. The top-10 spread is 19.0 points.

Compare top models

Gemini 3.5 Flash vs GPT-5.5 GPT-5.5 vs Kimi K3 Kimi K3 vs GLM-5.2 GLM-5.2 vs GPT-5.4

About APEX-Agents-AA

Definition and scoring

Organisation: Artificial Analysis / Mercor
Category: Agents
Version: Not evaluated
Direction: higher is better
Ranking use: Weighted 1.0%
Contamination risk: Unknown

Open benchmark source ↗

FAQ

Quick answers

What does APEX-Agents-AA measure?

Long-horizon professional multi-application agent benchmark implemented independently by Artificial Analysis.

Which model scores highest on APEX-Agents-AA?

Gemini 3.5 Flash by Google currently leads with 47.1%.

How many models are evaluated on APEX-Agents-AA?

24 models in the LuminaBench cohort have a qualifying score on this benchmark.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 1.0% of its capability category. Gemini 3.5 Flash is overall #17.

Terminal-Bench

2.1 · 80 results · Weighted 1.6%

Berkeley Function-Calling Leaderboard

V3 · 2 results · Not Weighted

τ-bench

Rolling · 37 results · Weighted 1.6%

GAIA

Rolling · 13 results · Weighted 1.6%

BrowseComp

Rolling · 24 results · Weighted 1.6%

OSWorld

Rolling · 11 results · Weighted 1.6%

Leaderboard · 24 models

Gemini 3.5 Flash

GPT-5.5

Kimi K3

See the field, not just the table.

Benchmark score table (24 models)

Compare top models

Definition and scoring

Quick answers

What does APEX-Agents-AA measure?

Which model scores highest on APEX-Agents-AA?

How many models are evaluated on APEX-Agents-AA?

Does this affect overall Lumina rank?

More Agents benchmarks

Leaderboard · 24 models

Gemini 3.5 Flash

GPT-5.5

Kimi K3

See the field, not just the table.

Benchmark score table (24 models)

Compare top models

Definition and scoring

Quick answers

What does APEX-Agents-AA measure?

Which model scores highest on APEX-Agents-AA?

How many models are evaluated on APEX-Agents-AA?

Does this affect overall Lumina rank?

More Agents benchmarks