Expired50% confidenceBenchmarkExact time
Citing a third-party programming benchmark, OpenAI states that Sol scored 80 points, 2.8 points higher than Claude's comparable model.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
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
UnverifiedSol 在编程智能体指数全推理模式下拿到 80 分，刷新 SOTA，比 Faber-5 高 2.8 分72% similarUnverifiedCodex One's performance on SWE-bench is slightly better than O3 High, achieving an accuracy rate of 75% on OpenAI's internal software engineering tasks, compared to 70% for O3 High.65% similarUnverifiedIn the Tieba mini-site project evaluation, Sol ranked first with 83.95 points, GLM 5.2 ranked second with 76.59 points, Terra ranked third with 75.97 points, and Luna ranked fourth with 68.56 points.64% similarUnverifiedOpenAI o4 scored 68.8% on the ARC-AGI-2 test, and GPT-5.2 scored 52.9%.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/489450API
curl https://kongchang.com/api/v1/knowledge/claims/489450MCP
get_claim(id=489450)