Expired50% confidenceBenchmarkExact time
On the Terminal Bench 2.1 benchmark, Ultra mode approaches 92%, surpassing competitors' performance of 88%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/18/2026(expired)
Sources
Related Claims
Unverified当所有模型在某评测上都能拿90分时该评测就应退场，即基准饱和现象62% similarUnverifiedo1在国际数学奥林匹克级别问题上正确率超过80%，远高于GPT-4的约13%62% similarUnverifiedNational competition plagiarism standards: over 20% disqualifies a paper from provincial first prize or national first prize; over 40% results in no award; over 60% results in a public notice of criticism.62% similarUnverifiedIn a 100-question SWE-verified test, Rocky averaged approximately 80 points across two rounds, close to DeepSeek-V4-Pro's official score of 80+62% similarUnverifiedCompared to extra high, the Max level consumes up to twice as many tokens, but delivers only a 4% to 10% performance improvement across various benchmarks.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/572658API
curl https://kongchang.com/api/v1/knowledge/claims/572658MCP
get_claim(id=572658)