chiprook
← AI
AIOctober 12, 2026, 06:14

VCIS-Bench: Claude Sonnet 5.5 beats Gemini 3.7 Flash and Gemma 4 on spatial grid puzzles

A developer released VCIS-Bench, a synthetic 7×7 grid benchmark where an agent must grab a key and unlock a door without hitting walls. Claude Sonnet 5.5 scored 100% on all three tasks, Gemini 3.7 Flash 0.97, and open-weights Gemma 4 26B A4B 0.85.

VCIS-Bench: Claude Sonnet 5.5 beats Gemini 3.7 Flash and Gemma 4 on spatial grid puzzles
#Anthropic#Google#Claude#Gemini
Read next
AI

StepFun releases Step 3.7 Flash: 198B MoE with 11B active parameters

AI

Google launches Gemini 3.7 Flash for coding and agentic tasks

AI

Study: Claude Sonnet 4.6 and GPT-5.5 Show Gender Bias in Moral Judgments, DeepSeek V4-Flash Doesn't

AI

TaichuAI Open-Sources ZDTaichu5.0-9B Spatial Multimodal Model