Skip to content

Benchmark

GDPval-AA

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are Elo ratings. Higher ratings are better.

Models measured
27
Scores recorded
30
Sources
2
IQ weight
1.32

Score distribution

1,509.1
Mean
1,581
Median
234.5
Std dev
1,211
10th percentile
1,769
90th percentile
953
Min
9531,853
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.32 for chat models, measured over 27 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5.1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,853
2GLM 5.3 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,773
3GLM 5.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,769
4GPT 5.5Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,769
5Muse Spark 1.3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,754
6Claude Opus 4.7Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,753
7Claude Opus 4.7independent Artificial Analysis run · shell + web browsing· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,753
8Grok 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,753
9Hy4 preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,678
10Claude Sonnet 4.6Gemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,676
11Gemini 3.5 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,656
12Claude Sonnet 4.6220 tasks · shell + web browsing· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,633
13Inkling 2 Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,630
14Claude Opus 4.6Anthropic Opus 4.6 launch page · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,606
15MiMo V2.5 ProXiaomi MiMo-V2.5-Pro official launch table · provider-reported comparison table· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,581
16Gemini 3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,545
17Gemini 3.7 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,525
18MiniMax M2.7MiniMax M2.7 official release prose· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,495
19MiMo V2 ProXiaomi MiMo-V2-Pro release table · Artificial Analysis GDPVal-AA · snapshot/date unspecified· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,426
20Muse Spark Open· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,401
21DeepSeek V4 FlashDeepSeek V4 official model card · Think Max· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,395
22Gemini 3.1 Pro Previewsetting not named by the source· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,317
23Gemini 3.1 Pro PreviewGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,314
24Inkling Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,269
25Inkling FP4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,238
26Mercury 3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,237.2
27NVIDIA Nemotron 3.5 Super VL 120B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,211
28Gemini 3 FlashGemini 3.5 Flash Model Card · May 2026· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,204
29Ling 3.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,107
30Muse Glimmer 30B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim953

Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.