Epoch's SpireClimb Shows GPT-6 Astra Winning 98% of Slay the Spire Runs

Epoch AI's new SpireClimb benchmark shows frontier models leapt from near-zero win rates to 90%+ on base difficulty in six months.

Read the full story at AlphaSignal ↗

Timeline · 1 report

2026-10-14 19:59 · AlphaSignal
Epoch's SpireClimb Shows GPT-6 Astra Winning 98% of Slay the Spire Runs

More stories

OpenAI engineer arrested for allegedly planting surveillance device — Sky News Technology
Eight US states sue OpenAI over alleged safety risks to children — The Verge AI
Amp integrates OpenAI models to expand AI agent capabilities beyond coding — OpenAI News
Third Moment Research runs 10,000 agents for 36+ hours using OpenAI models — OpenAI News
CoreWeave releases tutorial for benchmarking GPT-6 Astra, Claude Fable 5.1, and GPT-5.6 Sol — CoreWeave Blog
Higgsfield expands AI video capabilities with new models — OpenAI News
AWS enables global cross-Region inference for OpenAI GPT models on Bedrock — AWS Machine Learning Blog
Five AI chatbots were tested on accuracy. The results weren’t good — The Independent Tech

Get the daily brief of stories like this at 6:30 every morning →