List your company

Competitive Programming on ICPC-Eval (Refine@5, #T)

48.3Refine@5Gemini-3-pro
1.60413.72725.8537.973May 23, 2026

Evaluation results

MethodLinks
Gemini-3-proBackbone=Gemini-3-pro2026.0548.31.4
o4-mini-highBackbone=o4-mini-high2026.0542.41.2
DeepSeek-V3.2-ReasonerBackbone=DeepSeek-V3.2-Reasoner2026.0533.91.6
CP-AgentBackbone=DeepSeek-V3.2-Chat2026.0533.91.3
ReflexionBackbone=DeepSeek-V3.2-Chat2026.0532.51.2
CodeSimBackbone=DeepSeek-V3.2-Chat2026.0529.71.2
AlphaCodiumBackbone=DeepSeek-V3.2-Chat2026.0529.71.7
ReflexionBackbone=Qwen3-235B-A22B-Instruct2026.0527.11.2
CP-AgentBackbone=Qwen3-235B-A22B-Instruct2026.0526.31.6
CodeSimBackbone=Qwen3-235B-A22B-Instruct2026.0525.42.1
AlphaCodiumBackbone=Qwen3-235B-A22B-Instruct2026.0525.41.6
DeepSeek-V3.2-ChatBackbone=DeepSeek-V3.2-Chat2026.0522.91.2
Qwen3-235B-A22B-InstructBackbone=Qwen3-235B-A22B-Instruct2026.0521.21.4
DeepSeek-R1Backbone=DeepSeek-R12026.0514.42.1
ReflexionBackbone=GPT-4o2026.057.72.4
CP-AgentBackbone=GPT-4o2026.057.71.6
AlphaCodiumBackbone=GPT-4o2026.055.92.7
CodeSimBackbone=GPT-4o2026.055.11.5
GPT-4oBackbone=GPT-4o2026.053.41.8