Benchmark · Other
GDPval-AA v2.1
Elo (Elo) Higher is better Benchmark homepage
What it measures
GDPval-AA is Artificial Analysis' independent evaluation of models on OpenAI's GDPval dataset of economically valuable knowledge-work tasks. Version 2.1 uses 220 tasks across 44 occupations and 9 industries, which require outputs such as documents, slides, diagrams and spreadsheets. Models work in an agentic loop with shell access and web browsing (via the Stirrup harness). Blind pairwise comparisons by an LLM judge are aggregated into Elo ratings anchored to DeepSeek V4.1 Flash (max) at 1600.
Caveats:
Elo is a relative rating: values change as models are added and the scale is re-anchored, so always store the date a rating was read. For example, Anthropic's launch page gives 1846 for Claude Opus 5.5, while the leaderboard on 2026-10-12 shows 1866 (Max, Default Fallback). Earlier versions (Google cites "GDPVal-AA v2") are not comparable. Judging is done by an LLM, not by human experts. Anthropic models are listed with "Default Fallback", meaning refused steps are routed to a fallback model. AA-Briefcase and GDP.pdf are separate evaluations.
Results
| Model | Score | Setting | Reported by | Measured | Verified | Source |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 Anthropic | 1,866 Elo | max effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Muse Wonder Full Meta | 1,857 Elo | v2.1; max effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Opus 5.5 Anthropic | 1,846 Elo | max effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Opus 5.5 System Card (Anthropic, 2026-09-22) |
| Claude Sonnet 5.5 Anthropic | 1,844 Elo | max effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Sonnet 5.5 System Card (Anthropic, 2026-09-28) |
| Claude Sonnet 5.5 Anthropic | 1,838 Elo | max effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Opus 5.5 Anthropic | 1,836 Elo | xhigh effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Opus 5.5 Anthropic | 1,820 Elo | xhigh effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Opus 5.5 System Card (Anthropic, 2026-09-22) |
| Muse Wonder Meta | 1,770 Elo | v2.1; max effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Fable 5.1 Anthropic | 1,756 Elo | max effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| DeepSeek-V4.1-Flash DeepSeek | 1,746 Elo | snapshot DeepSeek-V4.1-Flash-1017; max effort; Stirrup agent loop (shell + web browsing) | Third party | — | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Fable 5.1 Anthropic | 1,735 Elo | max effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Opus 5.5 System Card (Anthropic, 2026-09-22) |
| Claude Fable 5.1 Anthropic | 1,733 Elo | xhigh effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Sonnet 5.5 Anthropic | 1,730 Elo | xhigh effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Sonnet 5.5 Anthropic | 1,725 Elo | xhigh effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Sonnet 5.5 System Card (Anthropic, 2026-09-28) |
| Grok 4.7 xAI | 1,714 Elo | v2.1; xhigh effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.7 xAI | 1,707 Elo | v2.1; high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Opus 5.5 Anthropic | 1,705 Elo | high effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.7 xAI | 1,695 Elo | GDPval, version not stated; xhigh effort | Self-reported | — | 2026-10-25 | Introducing Grok 4.7 |
| Muse Spark 1.3 Meta | 1,681 Elo | v2.1; max effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.8-Max Alibaba Cloud (Qwen) | 1,666 Elo | snapshot 'Qwen3.8 Max (0902)'; effort not shown; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.6 xAI | 1,647 Elo | v2.1; xhigh effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Fable 5.1 Anthropic | 1,633 Elo | high effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Muse Spark 1.3 Meta | 1,627 Elo | v2.1; xhigh effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 4 Argon Google | 1,624 Elo | v2.1; high effort; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Haiku 5.5 Anthropic | 1,620 Elo | max effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Haiku 5.5 System Card (Anthropic, 2026-10-07) |
| Claude Haiku 5.5 Anthropic | 1,618 Elo | max effort | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.6 xAI | 1,618 Elo | v2.1; high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.6 xAI | 1,613 Elo | v2.1; medium effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.8-2.4T-A95B Alibaba Cloud (Qwen) | 1,607 Elo | 'Qwen3.8 2.4T A95B'; effort not shown; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.6 xAI | 1,605 Elo | GDPval, version not stated; high effort | Self-reported | — | 2026-10-25 | Introducing Grok 4.7 |
| Qwen3.8-Max Alibaba Cloud (Qwen) | 1,604 Elo | 'Qwen3.8 Max'; effort not shown; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| DeepSeek-V4.1-Flash DeepSeek | 1,600 Elo | max effort; Stirrup agent loop (shell + web browsing) | Third party | — | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6.1 Sol OpenAI | 1,596 Elo | max effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Opus 5.5 Anthropic | 1,584 Elo | medium effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.7 xAI | 1,583 Elo | v2.1; low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Astra OpenAI | 1,574 Elo | max effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Astra OpenAI | 1,557 Elo | xhigh effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Sonnet 5.5 Anthropic | 1,550 Elo | high effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Fable 5.1 Anthropic | 1,545 Elo | medium effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6.1 Sol OpenAI | 1,539 Elo | xhigh effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Astra OpenAI | 1,525 Elo | high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6.1 Sol OpenAI | 1,512 Elo | high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Haiku 5.5 Anthropic | 1,511 Elo | xhigh effort | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Sol OpenAI | 1,509 Elo | max effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Astra OpenAI | 1,487 Elo | medium effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Muse Spark 1.2 Meta | 1,475 Elo | v2.1; xhigh effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Fable 5.1 Anthropic | 1,468 Elo | low effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Sol OpenAI | 1,457 Elo | xhigh effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6.1 Sol OpenAI | 1,450 Elo | medium effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| DeepSeek-V4-Pro DeepSeek | 1,449 Elo | max effort; Stirrup agent loop (shell + web browsing) | Third party | — | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.8 Flash Google | 1,436 Elo | v2.1; high effort; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Luna OpenAI | 1,433 Elo | max effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.8 Flash Google | 1,423 Elo | v2.1; medium effort; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Mistral Large 4 Mistral AI | 1,422 Elo | v2.1, 'Mistral Large 4 Preview'; effort not stated; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.8-27B Alibaba Cloud (Qwen) | 1,419 Elo | xhigh effort; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Haiku 5.5 Anthropic | 1,418 Elo | high effort | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.6 xAI | 1,409 Elo | v2.1; low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Muse Spark Open Meta | 1,401 Elo | v2.1; xhigh effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Sol OpenAI | 1,396 Elo | high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Astra OpenAI | 1,391 Elo | low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.5 xAI | 1,381 Elo | v2.1; high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.8-27B Alibaba Cloud (Qwen) | 1,380 Elo | medium effort; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Luna OpenAI | 1,363 Elo | xhigh effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.8-27B Alibaba Cloud (Qwen) | 1,363 Elo | low effort; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Sol OpenAI | 1,346 Elo | medium effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Luna OpenAI | 1,343 Elo | high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| DeepSeek-V4.1-Flash DeepSeek | 1,340 Elo | non-reasoning; Stirrup agent loop (shell + web browsing) | Third party | — | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Sonnet 5.5 Anthropic | 1,323 Elo | medium effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6.1 Sol OpenAI | 1,316 Elo | low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.8 Flash Google | 1,307 Elo | v2.1; low effort; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.6 Flash Google | 1,286 Elo | v2.1; high effort; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Haiku 5.5 Anthropic | 1,277 Elo | medium effort; run by Artificial Analysis | Self-reported | — | 2026-10-12 | Claude Haiku 5.5 System Card (Anthropic, 2026-10-07) |
| Claude Haiku 5.5 Anthropic | 1,274 Elo | medium effort | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Luna OpenAI | 1,263 Elo | medium effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Sol OpenAI | 1,251 Elo | non-reasoning | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Opus 5.5 Anthropic | 1,234 Elo | low effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Muse Spark 1.1 Meta | 1,211 Elo | v2.1; xhigh effort; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Sol OpenAI | 1,196 Elo | low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Sonnet 5.5 Anthropic | 1,175 Elo | low effort; Default Fallback | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Haiku 5.5 Anthropic | 1,122 Elo | low effort | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.8-27B Alibaba Cloud (Qwen) | 1,074 Elo | non-reasoning; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| DeepSeek-V4-Pro DeepSeek | 1,056 Elo | non-reasoning; Stirrup agent loop (shell + web browsing) | Third party | — | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Luna OpenAI | 1,045 Elo | non-reasoning | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| GPT-6 Luna OpenAI | 1,032 Elo | low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.5 Flash-Lite Google | 985 Elo | v2.1; effort not shown; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.3 xAI | 941 Elo | v2.1; non-reasoning | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Grok 4.3 xAI | 938 Elo | v2.1; high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Claude Sonnet 4.5 Anthropic | 909 Elo | Reasoning | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.1 Pro Google | 791 Elo | v2.1; effort not shown; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Muse Glimmer Meta | 783 Elo | v2.1; high reasoning; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Qwen3.7-Plus Alibaba Cloud (Qwen) | 767 Elo | 'Qwen3.7 Plus'; effort not shown; Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Mistral Medium 3.5 Mistral AI | 761 Elo | v2.1; effort not stated; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemma 4 31B Google | 619 Elo | v2.1; reasoning; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| gpt-oss-120b OpenAI | 608 Elo | high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemma 4 31B Google | 547 Elo | v2.1; non-reasoning; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Command A+ Cohere | 514 Elo | GDPval-AA v2.1; effort not stated (row label 'Command A+'); Stirrup agent harness with shell and web browsing | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Gemini 3.1 Flash-Lite Google | 430 Elo | v2.1; effort not shown; Artificial Analysis Stirrup agent harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Mistral Large 3 Mistral AI | 420 Elo | v2.1; no effort label; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Amazon Nova 2 Lite Amazon | 366 Elo | v2.1; high effort; Stirrup agent harness (shell + web browsing) | Third party | 2026-10-12 | 2026-10-12 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Mistral Small 4 Mistral AI | 362 Elo | v2.1, 'Mistral Small 4 (Reasoning)'; reasoning on, effort level not stated; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| gpt-oss-20b OpenAI | 339 Elo | high effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| North Mini Code Cohere | 311 Elo | GDPval-AA v2.1; effort not stated (row label 'North Mini Code'); Stirrup agent harness with shell and web browsing | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Ministral 3 14B Mistral AI | 247 Elo | v2.1; no effort label; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Ministral 3 8B Mistral AI | 213 Elo | v2.1; no effort label; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| gpt-oss-120b OpenAI | 171 Elo | low effort | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Ministral 3 3B Mistral AI | 29 Elo | v2.1; no effort label; Stirrup harness (shell + web browsing) | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Llama 4 Scout Meta | -164 Elo | v2.1; no effort setting; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
| Llama 4 Maverick Meta | -276 Elo | v2.1; no effort setting; Artificial Analysis Stirrup harness | Third party | 2026-10-25 | 2026-10-25 | GDPval-AA v2.1 Leaderboard (Artificial Analysis) |
Sorted by score, without rank numbers: each result was measured under its own setting (effort level, harness, tools, test version), as stated by the source. Scores are only directly comparable between rows with the same setting. Self-reported results come from the provider and have not been reproduced independently.