PCFBench - Background Mapping: leaderboard
Metric: Exact-match accuracy (%; mapping 109 material names to the correct ecoinvent activity; agentic harness; from PCFBench, a product-carbon-footprint benchmark of 614 expert-labelled items; scored deterministically or by a temperature-0 judge; higher is better). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 93 |
| 2 | Claude Opus 4.6 | 85 |
| 3 | DeepSeek V3.2 | 84 |
| 4 | Claude Sonnet 4.6 | 82 |
| 5 | GPT-5.5 | 80 |
| 6 | Claude Haiku 4.5 | 79 |
| 7 | GPT-5.4 Mini | 75 |
| 8 | Gemini 3 Flash | 68 |
Interactive version: theaggregate.ai/benchmark?slug=pcfbench-background-mapping · How It Works · Data refreshed daily, snapshot 2026-09-26.