Unverified50% confidenceBenchmarkExact time
DeepSeek Coder V1 33B Instruct版本在HumanEval Python代码生成测试中通过率达79.3%，超越GPT-3.5 Turbo
1
Sources
50%
Confidence
Long-term
Relevance
7/10/2026
First Seen
Sources
DeepSeek Coder Deep Dive: How This Open-Source Code LLM Outperforms GPT-4
bilibiliAI论文许老师教授7/8/2026
Related Claims
VerifiedThe open-source version of DeepSeek-Coder includes multiple parameter scales such as 1.3B, 6.7B, and 33B.75% similarVerifiedDeepSeek performs excellently on multiple code benchmarks (such as HumanEval, MBPP, and LiveCodeBench), with some metrics approaching or even surpassing GPT-4-level models.74% similarUnverifiedDeepSeek-Coder-V2 has long ranked among the top on code evaluation leaderboards such as HumanEval.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/468000API
curl https://kongchang.com/api/v1/knowledge/claims/468000MCP
get_claim(id=468000)