Unverified50% confidenceEventExact time
In the first round of the PPO reinforcement learning task, Opus chose to train a quadruped ant for nearly an hour, 75 million steps, scoring 7254; Kimi chose to train a bipedal cheetah for less than six minutes, barely passing with 3000 points.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/25/2026
First Seen
Valid until: 11/8/2026
Sources
Kimi K3.1 Pro vs Opus 5.5 in 13 Long-Running Tasks: Which One Is Actually Worth Using?
youtubePrompt Engineering10/25/2026
Related Entities
Related Claims
UnverifiedTraining uses batch size 256, 400,000 steps, EMA 0.9999, linear learning rate decay over the final quarter, and torch.compile achieves a 2.4× speedup over eager mode.71% similarUnverifiedTraining a quadruped robot walking policy on Isaac Gym can converge in as little as approximately 20 minutes, whereas traditional methods may require several hours.66% similarUnverifiedA single Mario training run of 25 million steps requires more than two days just for the simulation phase.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1097355API
curl https://kongchang.com/api/v1/knowledge/claims/1097355MCP
get_claim(id=1097355)