RepoRef: leaderboard

Metric: Success rate (%; exact match of the submitted GitHub issue, pull request or commit URL against the hidden target over the 400 developer-chat segments from 92 repositories, each with one indirect reference to resolve; a ReAct agent with 22 read-only GitHub tools and a budget of 10 tool calls, provider-default model settings, or the Claude Code harness through a read-only GitHub MCP server). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.

Top models

#ModelScore
1Gemini 3 Flash (Preview)67
2DeepSeek V4 Pro60.25
3Grok 4.1 Fast (Reasoning)54
4GLM-5.152
5Claude Sonnet 4.651.25
6GPT-5 Mini49
7Gemini 2.5 Flash Lite4

Interactive version: theaggregate.ai/benchmark?slug=reporef · How It Works · Data refreshed daily, snapshot 2026-09-26.