Basically, it handles in-context learning by updating a recurrent memory when it sees a new task. When it comes time to solve the actual query, it iterates internally. The critical takeaway here is that intermediate reasoning states are never decoded into language. It's essentially "silent" reasoning.
How the latent workspace actually functions #
Unlike typical LLM agents that rely on prompt engineering to force a step-by-step verbal breakdown, BDH-CQ integrates memory and inference into the same computational fabric. The process looks like this:
-
Memory Update: The model takes in demonstrations of a task it has never seen before. Instead of just storing these in a KV cache, it updates its recurrent memory.
-
Latent Iteration: The query is processed through iterative computation. It cycles through the latent space to refine the answer.
-
Direct Output: It jumps straight to the solution without verbalizing the "scratchpad" steps.
This architecture avoids the need for task identifiers or specific demonstration pairs during the training phase. Even more impressive is that no parameters are updated during inference; it's all happening through the recurrent state.
Performance and Efficiency #
The numbers on this are wild when you consider the model size. A 150M-parameter setup—which is tiny compared to the behemoths we usually see—hit a 29.5% pass@2 on ARC-AGI-1.
Model Size: 150M parametersARC-AGI-1 Pass@2: 29.5%Cost per task:$0.00070
Comparing this to larger models that try to brute-force AGI benchmarks with massive prompt windows, BDH-CQ proves that recurrent latent states can be way more efficient than token-heavy reasoning. It's a real-world example of how moving away from "thinking out loud" might actually lead to better generalization on abstract reasoning tasks.
If you're looking for a deep dive into the math, the full paper is available here:
https://arxiv.org/abs/2608.09888
A 150M parameter model hitting 29.5% on ARC-AGI-1 is insane 8h ago
BDH-CQ is hitting tasks at $0.007 and that crushes OpenAI Luna 1d ago
Pathway's 150M model just hit 29. 3d ago
Next Steam catapults might be making a comeback on US aircraft →
a practical ChatGPT prompt guide, with plenty of directly applicable cases.