Agentic benchmark
APEX-Agents-AA leaderboard
Every model the catalog carries a published APEX-Agents-AA value for, ranked by that value.
CategoryAgentic
MeasurePass@1
Tasks452 professional-services agent tasks
DifficultyLong-horizon workplace agent tasks
Published byAPEX-Agents-AA Benchmark Leaderboard
Artificial Analysis' implementation of the APEX-Agents benchmark for long-horizon professional-services agent tasks.
APEX-Agents-AA ranking
26 models with a published APEX-Agents-AA value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Pass@1 |
|---|---|---|---|
| 1 |  Gemini 3.5 Flash | 47.1 | |
| 2 |  Kimi K3 | Moonshot AI | 41.3 |
| 3 |  GPT-5.6 Terra | OpenAI | 38.9 |
| 4 | GPT-5.5 | OpenAI | 37.7 |
| 5 | GPT-5.6 Luna | OpenAI | 35.8 |
| 6 |  GLM-5.2 | Z.AI | 33.7 |
| 7 | GPT-5.4 | OpenAI | 33.3 |
| 8 |  Claude Opus 4.6 (Adaptive) | Anthropic | 33.0 |
| 9 | Gemini 3.1 Pro | 32.0 | |
| 10 | APApodex 1.1 | Apodex | 31.2 |
| 10 | APApodex 1.1 Mini | Apodex | 31.2 |
| 12 | Kimi K2.6 | Moonshot AI | 28.5 |
| 13 | GPT-5.4 mini | OpenAI | 28.2 |
| 14 | GPT-5.4 nano | OpenAI | 24.9 |
| 15 |  DeepSeek V4 Pro 0813 | DeepSeek | 24.3 |
| 16 |  Qwen3.7 Plus | Alibaba | 22.4 |
| 17 |  Grok 4.3 | xAI | 17.0 |
| 18 |  Step 3.7 Flash | StepFun | 14.8 |
| 19 | GLM-5 | Z.AI | 14.5 |
| 20 | Kimi K2.5 | Moonshot AI | 11.5 |
| 20 | Kimi K2.5 (Reasoning) | Moonshot AI | 11.5 |
| 22 |  MiniMax M2.7 | MiniMax | 10.6 |
| 23 | GPT-OSS 120B | OpenAI | 3.1 |
| 24 |  MiMo-V2.5-Pro | Xiaomi | 2.4 |
| 25 |  Nemotron 3 Super 100B | NVIDIA | 1.8 |
| 26 | GPT-OSS 20B | OpenAI | 0.7 |
Evidence key: Observed
Rows are ordered by the value APEX-Agents-AA Benchmark Leaderboard published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.