Unverified50% confidenceBenchmarkExact time
In the Sokoban test, Gemini 4 Hydrogen scored 43 points, DeepSeek completed only two levels, and Kimi K3 completed only one level.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/14/2026
First Seen
Valid until: 2/12/2027
Sources
Gemini 4 Hydrogen Hands-On Test: Has Google Finally Caught Up in Coding?
bilibili程序员阿江-Relakkes11/13/2026
Related Entities
Related Claims
UnverifiedGemini 4 Hydrogen scored 57% on the Terminal Bench 4.1 test, more than three times the previous generation.69% similarUnverifiedOn the Artificial Analysis Intelligence Index, Gemini 4 Hydrogen scores 51 at the default high thinking level and can reach 54 at the highest xhigh thinking setting.68% similarUnverifiedGemini 4 Hydrogen scores 38.2% on the FrontierCode 2.0 benchmark.66% similarUnverifiedIn deep software engineering tests, Mistral Large 4 is inferior to Kimi K3, with a clear gap.65% similarUnverified在南宋临安追杀动作游戏测试中,DeepSeek V4 Pro的完成度超过Kimi K3,实现了可玩的MVP62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1171392API
curl https://kongchang.com/api/v1/knowledge/claims/1171392MCP
get_claim(id=1171392)