Expired50% confidenceOpinionExact time
HY3 has not yet displaced the open-source benchmark GLM 5.2, but trades wins and losses with DeepSeek V4 Pro on programming and agent benchmarks.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/13/2026
First Seen
Valid until: 10/11/2026(expired)
Sources
Related Claims
VerifiedDeepSeek-V3 outperformed open-source models such as Qwen2.5 and Llama 3.1 in multiple benchmarks.77% similarVerifiedDeepSeek released the V3.2 model, which matches Gemini 3.0 Pro on multiple benchmarks73% similarUnverifiedDeepSeek V4 comprehensively leads Qwen 3.5, Kimi K2.6, Gemma 5.1, and other open-source models across all metrics.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/495164API
curl https://kongchang.com/api/v1/knowledge/claims/495164MCP
get_claim(id=495164)