Partially Verified65% confidenceBenchmarkExact time
In the Go experiment, Astra wrote a Go program layer by layer in 336 lines of code and, with zero test runs, beat two smaller trained networks with a win rate above half; after 50 test-run self-play games, it beat an equally sized network trained on 20 million games.
3
Sources
65%
Confidence
Long-term
Relevance
11/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn experiment 5, against a model trained for 500 million steps, Astra's generated networks scored 20 times higher in Breakout, came close in Asterix, and reached 60% of the score in Sequest.75% similarUnverifiedIn the strictly comparable experiment with fifteen players and three impostors in the sixth league, the trained network's crew win rate over 300 rounds was 93.3%, while gpt-oss under the dialogue+intuition format won 16 of 20 rounds (80%), with a one-sided Fisher exact test p = 0.052.69% similarUnverifiedUsing the program's own gradients and the Adam optimizer, a Handwired twin network was trained on 2000 random orders; when only prices were changed, by step 100 it had exactly learned 549, 949, and 350 per kilogram, with zero error.66% similarUnverifiedAfter 52 minutes, the model correctly stated the game this network was trained to play: Breakout, with 88% confidence, without using any tools.66% similarUnverifiedWithout any training, Astra modified only 0.016% of the weights of a small model, raising its score on the Math 500 benchmark from 7.9% to 48.1%.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1135405API
curl https://kongchang.com/api/v1/knowledge/claims/1135405MCP
get_claim(id=1135405)