Expired50% confidenceOpinionExact time
Sakana claims that Fugu Ultra is on par with Claude Opus 5 across seven benchmarks, but actual testing indicates these claims are overly optimistic.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Claude Sonnet 5 Imminent, GPT-4.6 This Week? A Wave of Frontier Model Updates
bilibiliNixAI6/22/2026
Related Claims
UnverifiedThe test chose Claude Sonnet 4.6 rather than Opus 4.7 as the baseline, on the grounds that Opus is too strong to provide a meaningful comparison.62% similarUnverifiedOpus 5在AI Omniscience测试拿了31分，避免瞎编的能力优于Opus 4.861% similarVerifiedGemini 3.1 Pro achieved 77% on the advanced reasoning benchmark, while Claude Opus 4.6 scored 68%.61% similarUnverifiedClaude Opus 4.8在AI Coding基准测试中表现超越GPT-5.560% similarUnverifiedGLM 5.2 scored 74.4% on the Frontier SW benchmark, approaching the level of Anthropic Opus 4.8.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/40942API
curl https://kongchang.com/api/v1/knowledge/claims/40942MCP
get_claim(id=40942)