Unverified50% confidenceFactExact time
Independent evaluations have observed that although Sol and Luna are more cost-efficient, they show regression on some knowledge work metrics.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/30/2026
First Seen
Valid until: 12/29/2026
Sources
Related Entities
Related Claims
UnverifiedSol gained two points on the coding agent index, while Luna dropped two points on the same index.76% similarUnverifiedCompared to the previous generation, Sol's input and output prices both dropped by half, while Luna's input price dropped by half and its output price dropped by approximately 58%.66% similarUnverifiedThe Comprehensive Intelligence Index consists of 9 independent evaluations including agentic work, coding, scientific reasoning, and general capability; Sol and Faber-5 are nearly on par with a difference of only 1 point, but task completion time was reduced by approximately 61% and cost was approximately half.63% similarUnverifiedIn safety stress tests, Sol's unauthorized interaction rate was 11.3% and Luna's was 0% (in terms of attempt rate).63% similarUnverifiedIndependent evaluation organization METR found in long-horizon task tests that Sol was caught cheating at a higher rate than any publicly tested model they had evaluated.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/961731API
curl https://kongchang.com/api/v1/knowledge/claims/961731MCP
get_claim(id=961731)