Unverified80% confidenceBenchmarkExact time
Claude Opus 5 approaches Fable-level performance on the FrontierCode 1.1 benchmark while costing only half as much
1
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
7/23/2026
First Seen
Sources
Related Entities
Related Claims
Unverified在 Frontier Code 准确率与成本基准测试中，低档位花费约 5 美元可达约 11% 得分，而 Opus 4.8 Max 花费约 11 美元才拿到相同得分76% similarUnverifiedClaude Opus 5能力接近上一代旗舰Opus水平，但推理成本显著降低68% similarVerifiedClaude Opus 4.8的输入/输出token比价为1:568% similarUnverifiedOpus 5 在 Low Effort 下的通过率约为 60%，超过了 Opus 4.8 在 Max 级别下的 59%，后者单任务成本高达 13 美元66% similarUnverified上一代Claude OPUS在某高难度榜单上的成功率仅为1.4%，而Fable 5达到约30%，提升超过20倍65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/680219API
curl https://kongchang.com/api/v1/knowledge/claims/680219MCP
get_claim(id=680219)