Unverified50% confidenceFactExact time
In the tests, each Claude model (Haiku 4.5, Haiku 5.5, Sonnet 5.5, Opus 5.5) exhibited outcome-bias-induced score leakage on at least one case.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
UnverifiedThe two-stage classifier runs on top of Claude Sonnet 5.5, Opus 5 and 5.5, and Fable 5 and 5.1, and Anthropic says its false positive rate is below 0.1%.73% similarUnverifiedClaude Haiku 4.5 and Haiku 5.5 both scored the same 0.875 vs 0.875 on the two tasks, and the newer small model did not bring a score improvement.73% similarUnverifiedThe benchmark covers 9 models from 6 labs, including Qwen3 235B, Claude Haiku 4.5, Claude Opus 5.5, Gemini series, GPT-5.4 series, GLM-5, and Grok 4.20, of which 4 are reasoning models.70% similarUnverifiedIn benchmark tests, official demonstrations show Jev on par with Luna, Tara, and Sonnet 5—and even surpassing Opus 5—on metrics such as type safety, but at a fraction of the price.68% similarUnverifiedThe blogger judges that Haiku 5.5's absolute output quality does not match flagships such as Sonnet, Opus, or GPT-6 Astra, but delivers capability close to the frontier level of half a year ago at one-quarter of the price.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1043792API
curl https://kongchang.com/api/v1/knowledge/claims/1043792MCP
get_claim(id=1043792)