Unverified50% confidenceBenchmarkExact time
HY3's performance in physics simulation scenarios is roughly on par with Gemini 3.5, but at approximately 35 times lower cost.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/24/2026
First Seen
Valid until: 10/22/2026
Sources
腾讯HY3开源模型深度实测：轻量MoE硬刚旗舰级性能
bilibiliAI_Express7/8/2026
Related Claims
UnverifiedHY3 still lags behind GLM 5.2 (the current top-level open-source model) in overall performance.63% similarUnverifiedTencent positions HY3 as the strongest model with fewer than 300B parameters63% similarUnverified据视频对比数据，V4 Pro在智能水平上与GLM 5.2相当且价格略低59% similarUnverifiedTencent officially claims that Hunyuan3's performance is comparable to flagship models with 2 to 5 times the parameter scale.57% similarVerifiedMid-tier lightweight models (such as Claude 3 Haiku and GPT-4o-mini) are priced at approximately 1/10 to 1/20 of flagship models.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/606578API
curl https://kongchang.com/api/v1/knowledge/claims/606578MCP
get_claim(id=606578)