Enter opens the highlighted result. Esc closes.

GDPval-AA v2 leaderboard

As of November 18, 2026, among the results listed here, Anthropic Claude Fable 5.1 has the highest reported GDPval-AA v2 score, 1853, published by Anthropic, ahead of Anthropic Claude Opus 5 at 1824. 21 models have a GDPval-AA v2 result in our records.

Scores as the model makers published them, not re-run by us; each row links to its announcement.

Best reported score

1853

Anthropic Claude Fable 5.1

Reported by Anthropic, June 9, 2026

Runner-up

1824

Anthropic Claude Opus 5

Reported by Anthropic, June 9, 2026

21 models ranked · 9 makers

Reported GDPval-AA v2 scores

GDPval-AA v2 scores reported by model makers
Rank Model Score Setting Reported by Date
1 Claude Fable 5.1 1853 Not published Anthropic June 9, 2026 (model release)
2 Claude Opus 5 1824 Not published Anthropic June 9, 2026 (model release)
3 GLM 5.3 1769 Not published Z.ai August 18, 2026 (model release)
4 GPT-5.6 Sol 1748 Not published OpenAI August 7, 2025 (model release)
5 Claude Fable 5 1723 Not published Anthropic June 9, 2026 (model release)
6 Kimi K3 1686 max; Elo Moonshot AI July 16, 2026 (model release)
7 Muse Spark 1.2 1631 Not published Meta August 21, 2026 (model release) · read from the maker's chart
8 N Nex-N2.5-Pro 1628 Not published Nex AGI September 8, 2026 (model release)
9 GPT-5.6 Terra 1593 Not published OpenAI August 7, 2025 (model release)
10 GPT-5.6 Luna 1592 Not published OpenAI August 7, 2025 (model release)
11 GLM 5.2 1508 Not published Z.ai August 18, 2026 (model release)
12 GPT-5.5 1494 Not published OpenAI August 7, 2025 (model release)
13 N Nex-N2.5-Mini 1446 Not published Nex AGI September 8, 2026 (model release)
14 Gemini 3.6 Flash 1421 Not published Google July 21, 2026 · read from the maker's chart
15 Muse Spark Open 1401 xhigh Meta October 15, 2026 (model release)
16 Muse Spark 1.1 1371 Not published Meta August 21, 2026 (model release) · read from the maker's chart
17 Gemini 3.5 Flash 1349 Not published Google July 21, 2026 · read from the maker's chart
18 T Inkling Small 1269 Not published Thinking Machines July 30, 2026 (model release)
19 T Inkling 1238 Not published Thinking Machines July 30, 2026 (model release)
20 Nemotron 3.5 Lightning 832 Not published NVIDIA August 11, 2026 (model release)
21 Nemotron 3 Super 746 Not published NVIDIA August 11, 2026 (model release)

How much to trust these numbers

Source
Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
Settings
Settings differ between makers. Check the Setting column before comparing two scores.
Spread
1853 to 746 across 21 models.
Freshness
Newest result published October 15, 2026; oldest August 7, 2025.
Charts
Some values were read from charts in the announcements. Two independent readings had to match exactly.

What GDPval-AA v2 measures

Models create professional deliverables such as documents, spreadsheets, and slides; scores rank their performance through blind pairwise comparisons. Source: GDPval-AA v2.1 Benchmark