Unverified50% confidenceBenchmarkExact time

Claude Sonnet 5.5 achieved a perfect score of 100% on all three tasks in VCIS-Bench, with a total score of 1.00.

1

Sources

50%

Confidence

Short-term (~14 days)

Relevance

10/12/2026

First Seen

Valid until: 10/26/2026

Sources

VLM Spatial Reasoning Benchmark: Which Model Wins on 7×7 Grid Puzzles?

Related Entities

Claude Sonnet 5.5model VCIS-Benchproduct

Related Claims

VerifiedClaude Sonnet 5.5 achieved 70.6% on the Terminal benchmark, compared to only 10.3% for the previous generation.66% similar VerifiedClaude 3.5 Sonnet surpassed a 49% solve rate on the SWE-bench software engineering benchmark.66% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/1000533

API

curl https://kongchang.com/api/v1/knowledge/claims/1000533

MCP

get_claim(id=1000533)