September real benchmark leaderboard: GPT-6 Astra trails Claude by 0.53 points

hammertime @hammertime_oneOriginally posted 2026.09.14

Open original on X

OpenAI and Anthropic are now separated by 0.53 points on the leaderboard. top 5 AI models by real benchmark score - September 2026 > Claude Fable 5.1 - 84.61, $10/$50 per M tokens > GPT-6 Astra - 84.08, $10/$50 per M tokens > Claude Opus 5 - 81.97, $5/$25 per M tokens > Claude Fable 5 - 81.43, $10/$50 per M tokens > GPT-5.6 Sol - 80.63, $4/$20 per M tokens (cheapest overall) honestly it's kind of f*cking insane how fast "state of the art" turns into "last month's model" now

This post quotes another post: “September 2026 leaderboard top five: Astra ranks second at 84.08”