Skip to content
Agents tracked: 502 Releases (7d): 613 Agent status: 4 with issuesSince last limit reset: Claude 7d · Codex 5d For AI agents: Markdown Updated Nov 18, 2026

GPT-6 Astra leads Gid Bench with 73.9; Gemini 4 Argon and Claude Opus 5.5 tie for second

Average of 4 groups of 13 public benchmarks, 0 to 100. Live chart from AgentGid, updated daily.

GPT-6 Astra leads Gid Bench with 73.9; Gemini 4 Argon and Claude Opus 5.5 tie for second

Average of 4 groups of 13 public benchmarks, 0 to 100

0255075100GPT-6 Astra (OpenAI): 73.9GPT-6 Astra73.9Gemini 4 Argon (Google): 72.4Gemini 4 Argon72.4 = · 1 test fewerClaude Opus 5.5 (Anthropic): 72.3Claude Opus 5.572.3 =GPT-6.1 Sol (OpenAI): 71.2GPT-6.1 Sol71.2 = · 1 test fewerGPT-6.1 Terra (OpenAI): 70.3GPT-6.1 Terra70.3 = · 2 tests fewerClaude Sonnet 5.5 (Anthropic): 66.2Claude Sonnet 5.566.2 =Claude Opus 5 (Anthropic): 65.7Claude Opus 565.7 = · 1 test fewerClaude Fable 5 (Anthropic): 59.4Claude Fable 559.4 · 1 test fewerGPT-6 Sol (OpenAI): 57.7GPT-6 Sol57.7 · 1 test fewerGemini 3.7 Flash (Google): 52.1Gemini 3.7 Flash52.1 = · 2 tests fewerClaude Haiku 5.5 (Anthropic): 51.8Claude Haiku 5.551.8 = · 1 test fewerGemini 3.8 Flash (Google): 50.0Gemini 3.8 Flash50.0 = · 1 test fewerGrok 4.6 (xAI): 49.2Grok 4.649.2 = · 2 tests fewerKimi K3 (Moonshot): 46.2Kimi K346.2 = · 1 test fewerDeepSeek V4.1 Flash (DeepSeek): 45.9DeepSeek V4.1 Flash45.9 = · 2 tests fewer
Each model's best effort level. Scores less than 1 point apart are a tie (=). A model with fewer tests is marked with how many it lacks; prelim. = uses a maker's own number. Source: Gid Bench by AgentGid from Epoch AI (CC BY 4.0) and makers' launch pages · Nov 17, 2026.

Put this chart on your site

The chart updates by itself. Please keep the link to the source.

More charts