MiMo Coding Bench

coding agentscode

Modality: text · Language: en · Multilingual: no · Max score: 1

MiMo Coding Bench evaluates coding-agent capabilities on software engineering tasks reported with the MiMo model family.

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: agents, code. Language: en.