Skip to content
Original
ResearchEpoch AI · X· @EpochAIResearch·· 2 days ago

3.5K views · 96 likes · 5 reposts (at collection)

Epoch AI’s SpireClimb benchmark shows frontier models winning up to 20% of hardest runs

Who
Epoch AI
What
introduced the SpireClimb benchmark for evaluating frontier AI models on Slay the Spire

What happened

Six months ago, frontier AI models could barely beat Slay the Spire. Now, according to our new SpireClimb benchmark, they win 90%+ of runs at the easiest difficulty, and up to 20% of runs at the hardest.

Links in the post: 1

The story so far

Epoch AI introduced SpireClimb to evaluate frontier AI models on Slay the Spire, reporting win rates above 90% at the easiest difficulty and up to 20% at the hardest. Epoch AI said frontier models could barely beat the game six months earlier. Epoch AI also said it had recorded more than a thousand runs of Slay the Spire and its sequel for SpireClimb and would publish them on YouTube over the coming weeks, with subtitles showing the models’ in-game thinking. The videos are reconstructed from model play data rather than recorded live. Epoch AI said it does not plan to add future models’ runs to the archive.

Keep reading

Latest daily · Issue 24NVIDIA and AI4Bharat Introduce Nemotron-pretraining-Indic-v1 and an Open Data Generation Pipeline

Trending now

  1. 1Meta Releases Muse Spark Open With Mac and Cloud Support, Topping US Open-Weight Index
  2. 2Google Launches Gemini 4 Argon Globally, Expands Paid Access as Benchmark Results Vary
  3. 3OpenAI Updates GPT-6 Astra Pro in ChatGPT With Parallel Agents and Faster Responses
  4. 4Meta’s New Business Terms Raise Dispute Over Website Crawling and AI Training Rights

Tomorrow at 08:00, a briefing like this arrives in your inbox.

The same kind of briefing as the one above: a few stories worth reading. Free. Mail starts after you confirm, and every letter has a one-click unsubscribe. Read a past issue · 2026-10-17