SaaS Pricing Accuracy 2026: LLM Benchmark Ground Truth and Leaderboard
收藏Zenodo2026-06-04 更新2026-06-05 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.20549253
下载链接
链接失效反馈官方服务：
资源简介：
Ground truth dataset and leaderboard for the SaaS Pricing Accuracy 2026 LLM knowledge benchmark.
Contains verified monthly pricing for 8 canonical software products across 5 categories,
human-verified against vendor pricing pages in June 2026.
Files:
saas-pricing-benchmark-2026.csv -- ground truth prices per product
llm-saas-pricing-leaderboard-2026.csv -- scores for 13 frontier models
Top results: GPT-5.5 / GPT-5.4 / GPT-5.4-mini lead at 87.5%. Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 3 Flash at 75.0%. DeepSeek R1 lowest at 25.0%.
Scoring: Score = (correct products / 8) x 100.
A product is correct if the reported price is within +/-15% of ground truth.
Full benchmark leaderboard: SaaS Pricing Accuracy 2026 on Kaggle
Pricing source: verified software pricing database -- 494 products (2026)
Methodology: benchmark scoring and evaluation criteria
提供机构：
Zenodo创建时间：
2026-06-04
