MHGraphBench: leaderboard
Large language models (LLMs) are increasingly used in the mental health domain, yet it remains unclear how well they capture related biomedical knowledge and how reliably they apply it to clinically salient structured judgments. Here, we present a knowledge-graph (KG)-grounded benchmark for assessing LLMs on mental-health entity recognition, relation judgment, and two-hop reasoning. The benchmark
Metric: Avg$_{All^\ast}$ (self-reported). Source: benchmarklist.com. Saturation forecast: Around April 2027. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 | 70.28 | #248 |
| 2 | GPT-5.2 | 69.32 | #110 |
| 3 | GPT-4o | 69.1 | #343 |
| 4 | GPT-5 Mini | 68.38 | #186 |
| 5 | GPT-5.1 | 68.33 | #135 |
| 6 | Qwen 2.5 32B | 56.09 | #546 |
| 7 | Qwen 2.5 7B | 43.61 | #922 |
| 8 | Mistral 7B | 42.1 | #1397 |
| 9 | Llama 3.1 8B | 32.54 | #1151 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=mhgraphbench · How It Works · Data refreshed daily, snapshot 2026-10-20.