Experts doubted a small neural net could beat Craftax without an LLM's knowledge.
@hyperparticle3.1K followers
273,315 views
16 Comments
Comments from the original post on X (Twitter)

@hyperparticle
Our meta-harness is built on top of trackinizer, our OSS graph DB, self-improving its research over time. It launched 4,805 experiments, each training in a few hours on 1 GPU. We recently released a new UI update, which is free to use:
20 likes

@hyperparticle
The final 15M parameter model demonstrates expert-level play, such as creating a little jail for enemies, or pre-firing when the enemies round corners. It even discovered that ice attacks beat fire, and fire attacks beat ice. And it learned this on its own, from scratch, a first
11 likes

@hyperparticle
rekursiv at every level: agents run the research, the learner restarts from its own best moments, and each generation's games train the model that teaches the next. 📖 Full story: 💬 Join our Discord:
15 likes

@hyperparticle
The "3 weeks" here is also a bit conservative. The campaign actually beat the game in a few days, the rest of the time was spent performing ablations to improve the score, and get the meta-harness to discover their ideas faster.
4 likes

@maxm_ideas
15M params cooking for a cent while GPT-6 burns $83 for 1% 💀
3 likes

@ta_eis_eauton
Is meta harness OSS?
2 likes

@hyperparticle
We OSS the engine powering it, trackinizer:
8 likes

@Taodav
it feels like we’ve lost the plot here? craftax is an RL benchmark designed to test general aspects of RL algorithms, not LLM designed task-specific policies
1 like

@hyperparticle
The improvements proposed by the agents were fairly general, not task specific. E.g., representing the input board as a 2D grid and using CNNs. The agent mentioned it was inspired was inspired by works like VARC
1 like

@Michaelz1914
sick
1 like

@AndrewPrifer
Congrats on the release, I’ll have to play with this
1 like

@AlberFuen
Starcraft skirmish bot dev by agents seemed promising; but if i get It, this is autoresearch from @karpathy training a neural networks on the Game env: proposing new experiments, checking the loss... Seems promising and a great way to form factor universal recursive intelligences
1 like

@AlbertBuchard
RSI babay

@dfacto82
We should clarify this: the craftax meta-harness and results were achieved in a few days. The three weeks was spent improving the trackinizer system.

@MooncastOnline
Are you using a JEPA-like predictive coding world model + action controller model? I did something very basic like this a few months ago, my model was evolutionary and only like 1000 params though.

@kugelbl81659324
i was waiting for the wave of persistant agents just to see people using robotics ai primitives for the digital work (small high frequency models that don't use words) inside a bigger asyncronous harness