Unverified50% confidenceBenchmarkExact time
Gemini 4 set a new record on DeepSui with 77.9% (vs. Opus 5.5's 74.2%), but scored only 55% on FrontierSui and ranked last on TerminalBench 4.0 with 57.4%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/4/2026
First Seen
Valid until: 10/18/2026
Sources
Related Entities
Related Claims
UnverifiedGemini 4 scored 68.9% on the VALS index, beating Fable 5.1 and GPT-6 Astra, and surpassing Opus 5.5 by several percentage points.79% similarUnverifiedOn Harvey's LegalAgent benchmark, Gemini 4 achieved 19.6%, more than three times the score of leader Fable 5.1.76% similarPartially VerifiedGemini 4 Argon achieved a new record of 77.9% on the Deep7e long-workflow software engineering benchmark, surpassing Claude Opus 5.5.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975135API
curl https://kongchang.com/api/v1/knowledge/claims/975135MCP
get_claim(id=975135)