HELM Classic - Entity Matching: leaderboard
The "Entity Matching" subtask of HELM Classic. A scenario from Stanford HELM Classic, the holistic evaluation suite scoring language models across accuracy, robustness and calibration on a broad set of tasks.
Metric: Exact Match (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 67 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Palmyra X (43B) | 96.59 | #1317 |
| 2 | Mistral-7B-v0.1 | 93.81 | #1326 |
| 3 | text-davinci-002 | 93.08 | #1225 |
| 4 | text-davinci-003 | 93.04 | #1146 |
| 5 | GPT-3.5 Turbo (0613) | 91.77 | #1118 |
| 6 | LLaMA-30B | 90.83 | #1423 |
| 7 | mpt-30B | 89.74 | #1476 |
| 8 | mpt-30B Instruct | 89.54 | #1470 |
| 9 | vicuna-13B-v1.3 | 87.91 | #1447 |
| 10 | falcon-40B Instruct | 87.21 | #1458 |
| 11 | LLaMA-65B | 86.08 | #1303 |
| 12 | LLaMA-13B | 85.92 | #1516 |
| 13 | alpaca-7B | 85.26 | #1611 |
| 14 | text-curie-001 | 85.2 | #1632 |
| 15 | Llama 2 7B | 85.07 | #1520 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-entity-matching · How It Works · Data refreshed daily, snapshot 2026-10-20.