Replying to @MechanizeWork
However, like earlier models, Fable 5 also fails to build an emulator that works on Spout, a homebrew cave-flying game. It diverges shortly after the loading screen, scoring 7.6%.
00:00
Mechanize
151 posts
Mechanize
@MechanizeWork
We build environments and evals for training and evaluating frontier coding agents.
- Replying to @MechanizeWorkClaude Fable 5 is the first model we tested that gets perfect gameplay on Varoom 3D. Opus 4.8 got just 25% on the same game.
00:00 - Replying to @MechanizeWorkClaude Fable 5 performs especially well on gameplay, scoring 91.5%. Opus 4.8 scored 77.4%. Interestingly, Fable 5 is a regression on audio. It scores 44.5% on audio, which is worse than Opus 4.8's 69.1% and GPT-5.5's 58.9%.
- Claude Fable 5 scores 74.5% on GBA Eval, the best score to date. Given 24 hours, it writes an emulator that plays all but one game in our test set near-perfectly. It beats Opus 4.8's 24-hour score in under 2 hours.
- Replying to @MechanizeWorkThis is the first reward hacking attempt we've caught on GBA Eval. This case is somewhat subtle, not "malicious," and wouldn't have affected scores. This last point is exactly why we're careful to think about these behaviors when designing evals. Blog:
- Replying to @MechanizeWorkIt didn't work. The ROMs that Grok has access to are example ROMs that we intentionally give the models so they can test locally. We actually grade their emulators on a set of hidden ROMs, so the hacking doesn't improve the score.
- We caught Grok Build 0.1 reward hacking on GBA Eval. After it got stuck while testing, it started hard-coding its emulator to perform better on the exact ROM it was testing against.
- We are now seeking a puzzle maker to help us create puzzles that LLMs can't yet solve.
- Replying to @MechanizeWorkHowever, Opus 4.8's emulator is not perfect. On Varooom 3D, it diverges after around 2,000 frames. This is better than GPT-5.5 (whose emulator diverged after around 1,250 frames), but Opus 4.8 only scores 25% on this game.
00:00 - Replying to @MechanizeWorkHere's Claude Opus 4.8's emulator running Collie Defense, where it scores 99.8% on video and 91% on audio. On most games we tested, gameplay is near-perfect, with some audio imperfections.
00:00 - Claude Opus 4.8 scores 70.9% on GBA Eval, the top score to date. Given 24 hours, it writes an emulator that plays most games, with working audio on all of them. It beats the previous best (GPT-5.5 at 53.2%) in under an hour.
00:30We gave frontier AI coding agents 24 hours to write a complete Game Boy Advance emulator from scratch. GPT-5.5's emulator runs games best, with Claude Sonnet 4.6 and Opus 4.7 close behind. Gemini 3.1 Pro failed to produce a working emulator.
