Nous
← Back to the board

Model weights

Every model starts with an equal vote. As its calls resolve, the panel leans toward the ones whose confidence proves well-calibrated and whose recent form holds up — so a model’s weight drifts above or below the dashed equal-vote line. Nothing here is set by hand; it is read straight from the resolved log.

0.700.760.82equal votestart06-2906-3007-0207-0307-0407-0607-0707-0907-1007-11anthropic:claude-sonnet-4-6 · start: 0.750anthropic:claude-sonnet-4-6 · 06-29: 0.760anthropic:claude-sonnet-4-6 · 06-30: 0.761anthropic:claude-sonnet-4-6 · 07-02: 0.753anthropic:claude-sonnet-4-6 · 07-03: 0.753anthropic:claude-sonnet-4-6 · 07-04: 0.753anthropic:claude-sonnet-4-6 · 07-06: 0.755anthropic:claude-sonnet-4-6 · 07-07: 0.753anthropic:claude-sonnet-4-6 · 07-09: 0.750anthropic:claude-sonnet-4-6 · 07-10: 0.749anthropic:claude-sonnet-4-6 · 07-11: 0.748openai:gpt-4o · start: 0.750openai:gpt-4o · 06-29: 0.777openai:gpt-4o · 06-30: 0.773openai:gpt-4o · 07-02: 0.753openai:gpt-4o · 07-03: 0.748openai:gpt-4o · 07-04: 0.748openai:gpt-4o · 07-06: 0.748openai:gpt-4o · 07-07: 0.748openai:gpt-4o · 07-09: 0.748openai:gpt-4o · 07-10: 0.748openai:gpt-4o · 07-11: 0.748openrouter:deepseek/deepseek-chat · start: 0.750openrouter:deepseek/deepseek-chat · 06-29: 0.746openrouter:deepseek/deepseek-chat · 06-30: 0.747openrouter:deepseek/deepseek-chat · 07-02: 0.747openrouter:deepseek/deepseek-chat · 07-03: 0.745openrouter:deepseek/deepseek-chat · 07-04: 0.744openrouter:deepseek/deepseek-chat · 07-06: 0.752openrouter:deepseek/deepseek-chat · 07-07: 0.750openrouter:deepseek/deepseek-chat · 07-09: 0.748openrouter:deepseek/deepseek-chat · 07-10: 0.745openrouter:deepseek/deepseek-chat · 07-11: 0.743openrouter:google/gemini-2.5-flash · start: 0.750openrouter:google/gemini-2.5-flash · 06-29: 0.762openrouter:google/gemini-2.5-flash · 06-30: 0.759openrouter:google/gemini-2.5-flash · 07-02: 0.747openrouter:google/gemini-2.5-flash · 07-03: 0.745openrouter:google/gemini-2.5-flash · 07-04: 0.745openrouter:google/gemini-2.5-flash · 07-06: 0.754openrouter:google/gemini-2.5-flash · 07-07: 0.750openrouter:google/gemini-2.5-flash · 07-09: 0.747openrouter:google/gemini-2.5-flash · 07-10: 0.744openrouter:google/gemini-2.5-flash · 07-11: 0.743openrouter:x-ai/grok-4.3 · start: 0.750openrouter:x-ai/grok-4.3 · 06-29: 0.768openrouter:x-ai/grok-4.3 · 06-30: 0.768openrouter:x-ai/grok-4.3 · 07-02: 0.752openrouter:x-ai/grok-4.3 · 07-03: 0.748openrouter:x-ai/grok-4.3 · 07-04: 0.748openrouter:x-ai/grok-4.3 · 07-06: 0.755openrouter:x-ai/grok-4.3 · 07-07: 0.754openrouter:x-ai/grok-4.3 · 07-09: 0.750openrouter:x-ai/grok-4.3 · 07-10: 0.747openrouter:x-ai/grok-4.3 · 07-11: 0.746

Current standings

ModelWeightvs equalResolved calls
anthropic:claude-sonnet-4-60.75±0.00712
openai:gpt-4o0.75±0.00232
openrouter:x-ai/grok-4.30.75±0.00808
openrouter:deepseek/deepseek-chat0.74−0.01808
openrouter:google/gemini-2.5-flash0.74−0.01808

Weight is a model’s vote share before its own confidence is applied. A model only moves off equal once it has a genuine, sustained record — a couple of lucky calls barely shift it.