List your company

Code Generation on MBPP-ET

91.8Pass@1AgentCoder
11.232.12553.0573.975Dec 20, 2023May 2, 2024Sep 13, 2024Jan 25, 2025Jun 8, 2025Oct 20, 2025Mar 3, 2026

Evaluation results

MethodLinks
AgentCoderBackbone=GPT-42023.1291.875.9
AgentCoderBase Model=GPT-4, Evaluation Strategy=Agentic and Prompting Strategies2026.0391.8—
AgentCoderBackbone=GPT-4-turbo2023.1291.492.4
AgentCoderBackbone=GPT-4-Turbo2024.1291.4—
LCPBase Model=GPT-4, Evaluation Strategy=Agentic and Prompting Strategies2026.0389.7—
AgentCoderBackbone=GPT-3.5-turbo2023.1289.1142.1
AgentCoderBackbone=GPT-3.5-Turbo2024.1289.1—
AgentCoderBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0389.1—
LCPBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0386.4—
ThinkCoderBackbone=GPT-4-Turbo2024.1280.5—
ThinkCoder (GPT-4-Turbo)n=5, k=5, temperature=0.52024.1280—
ThinkCoder (GPT-4-Turbo)n=2, k=5, temperature=0.52024.1279.9—
ThinkCoder (GPT-4o)n=1, k=5, temperature=0.52024.1279.9—
ThinkCoderBackbone=GPT-4o2024.1279.2—
ThinkCoder (GPT-4o)n=2, k=5, temperature=0.52024.1279.2—
ThinkCoder (GPT-4o)n=5, k=5, temperature=0.52024.1279.2—
ThinkCoderBackbone=GPT-3.5-Turbo2024.1278.8—
ThinkCoder (GPT-4-Turbo)n=1, k=5, temperature=0.52024.1278.2—
GPT-4on=0, temperature=02024.1278.2—
AgentCoderBackbone=PaLM Coder2023.1275.5177.6
AgentCoderBackbone=Claude-instant-12023.1275.1277.4
ThinkCoder (CodeQwen1.5-7B-Chat)n=2, k=5, temperature=0.52024.1275—
ThinkCoder (CodeQwen1.5-7B-Chat)n=5, k=5, temperature=0.52024.1275—
ThinkCoder (Kimi)n=5, k=5, temperature=0.52024.1274.8—
ThinkCoder (CodeQwen1.5-7B-Chat)n=1, k=5, temperature=0.52024.1274.7—
ThinkCoder (Kimi)n=2, k=5, temperature=0.52024.1274.5—
GPT-4-Turbon=0, temperature=02024.1274.4—
ThinkCoder (Kimi)n=1, k=5, temperature=0.52024.1273.5—
CodeQwen1.5-7B-Chatn=0, temperature=02024.1267.8—
Kimin=0, temperature=02024.1267.4—
LPWBackbone=GPT-4o2024.1265.8—
LDBBackbone=GPT-4o2024.1265.4—
CodeCoRBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0365.2—
CodeCoTBackbone=GPT-3.5-turbo2023.126371.2
Self-CollaborationBackbone=GPT-42023.1262.119
Self-CollaborationBase Model=GPT-4, Evaluation Strategy=Agentic and Prompting Strategies2026.0362.1—
FlowGenscrum+TestLLM=GPT-3.5, Process Model=Scrum, Incorporates=CodeT2024.0358.7—
MapCoderLLM=GPT-42024.0557.7—
MapCoderBackbone=GPT-4-Turbo2024.1257.7—
CodeTLLM=GPT-3.52024.0356.7—
FlowGenscrumLLM=GPT-3.5, Process Model=Scrum2024.0356.7—
CoTLLM=GPT-42024.0556.2—
DirectLLM=GPT-42024.0554.7—
MapCoderLLM=ChatGPT2024.0554.4—
MapCoderBackbone=GPT-3.5-Turbo2024.1254.4—
MapCoderBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0354.4—
Self-debuggingBackbone=GPT-3.5-turbo2023.1252.342.1
GPT-4Backbone=GPT-4, Setting=Zero-Shot2023.1252.2—
GPT-4Evaluation Strategy=Zero-Shot2026.0352.2—
ReflexionLLM=GPT-3.52024.0352—
ReflexionLLM=GPT-42024.0551.9—
Self-PlanningLLM=GPT-42024.0550.4—
Self-collaborationLLM=ChatGPT2024.0549.5—
Self-CollaborationBackbone=GPT-3.5-turbo2023.1249.534.5
Few-ShotBackbone=GPT-3.5-turbo2023.1248.331.2
Few-ShotBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0348.3—
GPT-4-turboBackbone=GPT-4-turbo, Setting=Zero-Shot2023.1247.5—
ReflexionBackbone=GPT-3.5-turbo2023.1247.529.1
GPT-4-turboEvaluation Strategy=Zero-Shot2026.0347.5—
ReflexionBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0347.5—
ReflexionLLM=ChatGPT2024.0547.4—
INTERVENORBackbone=GPT-3.5-turbo2023.1247.128
RAPBackbone=GPT-3.5-turbo2023.1246.726.9
AnalogicalLLM=ChatGPT2024.0546.1—
ReActBackbone=GPT-3.5-turbo2023.1245.924.7
Self-EditBackbone=GPT-3.5-turbo2023.1245.924.7
ReActBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0345.9—
CodeX+CodeTBackbone=CodeX (175B), Setting=Augmented2023.1245.1—
Self-PlanningLLM=ChatGPT2024.0541.9—
Self-PlaningBackbone=GPT-3.5-turbo2023.1241.512.8
ToTBackbone=GPT-3.5-turbo2023.1240.810.9
AnalogicalLLM=GPT-42024.0540.3—
CoTLLM=ChatGPT2024.0539.6—
CodeXBackbone=CodeX (175B), Setting=Zero-Shot2023.1238.8—
DirectLLM=ChatGPT2024.0537.7—
GPT-3.5-turboBackbone=GPT-3.5-turbo, Setting=Zero-Shot2023.1236.8—
GPT-3.5-turboEvaluation Strategy=Zero-Shot2026.0336.8—
CoTBackbone=GPT-3.5-turbo2023.1234.8-5.4
CoTBase Model=GPT-3.5-turbo, Evaluation Strategy=Agentic and Prompting Strategies2026.0334.8—
StarCoderBackbone=StarCoder (15.5B), Setting=Zero-Shot2023.1233.4—
CodeGen-MonoBackbone=CodeGen-Mono (16.1B), Setting=Zero-Shot2023.1231.6—
ThinkCoder (LLama2-7B-Chat)n=5, k=5, temperature=0.52024.1229.8—
ThinkCoder (LLama2-7B-Chat)n=2, k=5, temperature=0.52024.1229.5—
ThinkCoder (LLama2-7B-Chat)n=1, k=5, temperature=0.52024.1229—
PaLM CoderBackbone=PaLM Coder, Setting=Zero-Shot2023.1227.2—
LLama2-7B-Chatn=0, temperature=02024.1222.9—
CodeGeeXBackbone=CodeGeeX (13B), Setting=Zero-Shot2023.1220.4—
Claude-instant-1Backbone=Claude-instant-1, Setting=Zero-Shot2023.1219.9—
Claude-instant-1Evaluation Strategy=Zero-Shot2026.0319.9—
IncoderBackbone=Incoder (6.7B), Setting=Zero-Shot2023.1214.3—
IncoderEvaluation Strategy=Zero-Shot, Model Parameters=6.7B2026.0314.3—