Six AI models played chess through their CLIs. None of them delivered checkmate.

Описание и применение

<p>I build SpaceApp, a self-hosted workspace for running AI coding CLIs side by side. On 18 October I let six models play chess against each other there, each through its own CLI. Nobody won by checkmate. The way they lost is the interesting part.</p> <h2> The setup </h2> <ul> <li> <strong>Players:</strong> Opus 5.5 and Fable 5.1 (through Claude Code), GPT-6-Astra and GPT-6.1-Sol (through Codex), all at xhigh reasoning; Gemini 4 Argon (High) and Gemini 3.8 Flash (High) (through the Gemini CLI).</li> <li> <strong>Format:</strong> a 4-round Swiss tournament, 12 games.</li> <li> <strong>Clock:</s

Задача

<p>I build SpaceApp, a self-hosted workspace for running AI coding CLIs side by side. On 18 October I let six models play chess against each other there, each through its own CLI. Nobody won by checkmate. The way they lost is the interesting part.</p> <h2> The setup </h2> <ul> <li> <strong>Players:</strong> Opus 5.5 and Fable 5.1 (through Claude Code), GPT-6-Astra and GPT-6.1-Sol (through Codex), all at xhigh reasoning; Gemini 4 Argon (High) and Gemini 3.8 Flash (High) (through the Gemini CLI).</li> <li> <strong>Format:</strong> a 4-round Swiss tournament, 12 games.</li> <li> <strong>Clock:</s

Результат

<p>I build SpaceApp, a self-hosted workspace for running AI coding CLIs side by side. On 18 October I let six models play chess against each other there, each through its own CLI. Nobody won by checkmate. The way they lost is the interesting part.</p> <h2> The setup </h2> <ul> <li> <strong>Players:</strong> Opus 5.5 and Fable 5.1 (through Claude Code), GPT-6-Astra and GPT-6.1-Sol (through Codex), all at xhigh reasoning; Gemini 4 Argon (High) and Gemini 3.8 Flash (High) (through the Gemini CLI).</li> <li> <strong>Format:</strong> a 4-round Swiss tournament, 12 games.</li> <li> <strong>Clock:</s

Доказательства

Six AI models played chess through their CLIs. None of them delivered checkmate.

Источник