Expired50% confidenceBenchmarkExact time
In the OS World computer operation test, Sol surpassed Claude's Opus series while consuming 85% less compute.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/10/2026(expired)
Sources
GPT-5.6 Released: Sol/Terra/Luna Three-Tier Model Family Takes On Claude Head-On
bilibiliAI技术投降派7/10/2026
Related Claims
UnverifiedOn the Artificial Analysis coding agent index, Sol set to the highest reasoning tier scored 80, surpassing Fable 5 by 2.8 points, while using less than half the output tokens, taking less than half the time, and costing roughly one-third less.69% similarUnverifiedOn SimpleBench, Sol Pro scored 71.7% and Sol approximately 65%, still 10 percentage points below Fable.60% similarUnverifiedThrough a series of strategy adjustments, the usage cost of Opus 5 can be reduced by up to 80%, while still outperforming Opus 4.8 on some benchmarks.60% similarUnverifiedSol achieved 92.2% on the web browsing benchmark BrowseComp and 62.6% on OSWorld 2.0, both current best results; on OSWorld it surpassed Opus 4.8 while reducing output tokens by 85%.60% 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.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/489451API
curl https://kongchang.com/api/v1/knowledge/claims/489451MCP
get_claim(id=489451)