GABench - Graph Retrieval - Node Level (Text-Paired Graphs): leaderboard
Metric: Task success rate (%; share of tasks whose final answer a DeepSeek-V4-Flash judge marks consistent with the verified ground truth; node-level graph retrieval (node attributes, neighbourhoods, edges, subgraphs and graph statistics) with its 10 retrieval tools on text-paired graphs (the HIV and PCBA molecule collections, 200 generated tasks per dataset and category); LLM agent in the OpenClaw harness, one run). Source: arxiv.org. Saturation forecast: Around December 2026. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 70.59 |
| 2 | GLM-5-Turbo | 60 |
| 3 | GPT-5 Mini | 50 |
| 4 | DeepSeek V4 Pro | 45.46 |
| 5 | GLM-4.5 Flash | 28.57 |
| 6 | Qwen 3 235B A22B | 0.43 |
Interactive version: theaggregate.ai/benchmark?slug=gabench-graph-retrieval-node-level-text-paired-graphs · How It Works · Data refreshed daily, snapshot 2026-09-29.