paper-with-me

Code Generation 벤치마크

Code Generation on CodeContests

11개 결과 · ⬇ CSV · JSON

Test Set pass@1

0.624 19.72 38.81 57.91 77 2023-06 2026-11 WizardCoder-15B — 1.11 (2023-06-14) CodeChain + WizardCoder-15B — 2.35 (2023-10-13) MoTCoder-15B — 26.34 (2023-12-26) MoTCoder-7B-v1.5 — 20.77 (2023-12-26) MapCoder (GPT-4) — 28.5 (2024-05-18) LPW (GPT-4o) — 34.7 (2024-11-21) CodeSim (GPT4) — 28.4 (2025-02-08) EG-CFG (DeepSeek-V3-0324) — 58.18 (2025-06-12) SEMAG — 3.3 (2026-03-16) SolidCoder — 77.0 (2026-04-20) MARS — 0.624 (2026-08-24) WizardCoder-15B — 1.11 (2023-06-14) CodeChain + WizardCoder-15B — 2.35 (2023-10-13) MoTCoder-15B — 26.34 (2023-12-26) MapCoder (GPT-4) — 28.5 (2024-05-18) LPW (GPT-4o) — 34.7 (2024-11-21) EG-CFG (DeepSeek-V3-0324) — 58.18 (2025-06-12) SolidCoder — 77.0 (2026-04-20)
RankModel Test Set pass@1Test Set pass@5Val Set pass@1Val Set pass@5 PaperCodeYear
1 SolidCoder 자동 추출 77.0––– SolidCoder: Bridging the Mental-Reality Gap in LLM Code Generation through Concrete Execution 2026
2 EG-CFG (DeepSeek-V3-0324) 58.18––– Execution Guided Line-by-Line Code Generation boazlavon/eg_cfg 2025
3 LPW (GPT-4o) 34.7––– Planning-Driven Programming: A Large Language Model Programming Workflow you68681/lpw 2024
4 MapCoder (GPT-4) 28.535.228.5– MapCoder: Multi-Agent Code Generation for Competitive Problem Solving md-ashraful-pramanik/mapcoder · Luoji-zju/Agents4PLC_release 2024
5 CodeSim (GPT4) 28.4––– CODESIM: Multi-Agent Code Generation and Problem Solving through Simulation-Driven Planning and Debugging kagnlp/CodeGenerator 2025
6 MoTCoder-15B 26.34–20.35– MoTCoder: Elevating Large Language Models with Modular of Thought for Challenging Programming Tasks dvlab-research/motcoder 2023
7 MoTCoder-7B-v1.5 20.77–16.72– MoTCoder: Elevating Large Language Models with Modular of Thought for Challenging Programming Tasks dvlab-research/motcoder 2023
8 SEMAG 자동 추출 3.3––– SEMAG: Self-Evolutionary Multi-Agent Code Generation 2026
9 CodeChain + WizardCoder-15B 2.353.292.483.30 CodeChain: Towards Modular Code Generation Through Chain of Self-revisions with Representative Sub-modules SalesforceAIResearch/CodeChain 2023
10 WizardCoder-15B 1.113.181.983.27 WizardCoder: Empowering Code Large Language Models with Evol-Instruct nlpxucan/wizardlm · nickrosh/evol-teacher · kyle-lyu/data-efficient-finetuning · +1 2023
11 MARS 자동 추출 0.624––– MARS: Multi-Specialist LLM Relay System for Competitive Programming 2026
1–11 / 11 페이지당 10 20 50 100