Verified70% confidenceFactTime unknown
Codex achieved a pass@1 accuracy of 28.8% on the HumanEval benchmark
4
Sources
70%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Related Entities
Related Claims
Unverified初版 Codex 采用约120亿参数规模，在 HumanEval 基准测试中达到 28.8% 的 pass@1 通过率83% similarUnverified早期Codex在HumanEval基准测试中以37%的准确率独立解决编程题，达到当时最高水平81% similarUnverifiedCodex于2021年发布时在HumanEval基准上达到28.8%的pass@1得分，而GPT-4在同一基准上已超过67%76% similarUnverifiedThe original Codex achieved a pass@1 score of 28.8% on the HumanEval benchmark, whereas using GPT-3 directly yielded only 0%.76% similarUnverifiedClaude 3.5 Sonnet在HumanEval基准测试上的得分超过90%，而2021年最初的Codex模型得分仅为28.8%64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/13593API
curl https://kongchang.com/api/v1/knowledge/claims/13593MCP
get_claim(id=13593)