Unverified50% confidenceBenchmarkExact time
DeepSeek-V3和Qwen2.5系列在代码和结构化数据生成的基准测试（如HumanEval、MBPP）中表现已接近GPT-4级别，但API调用成本仅为其1/10左右
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/17/2026
First Seen
Valid until: 10/15/2026
Sources
Cite This Claim
Stable URI
https://kongchang.com/claim/541211API
curl https://kongchang.com/api/v1/knowledge/claims/541211MCP
get_claim(id=541211)