JevBench vs PacBench
Both are Research & Data tools. Both are free.
Grok| JevBench | PacBench | |
|---|---|---|
| Tagline | A reproducible benchmark for typed decision models | How well can models one-shot a Pac-Man game? |
| Pricing | Free | Free |
| Categories | Research & Data, LLMs & Infrastructure | Coding & Dev Tools, Research & Data |
| Built for | Developers, Researchers | Developers |
| Platforms | Web | Web |
| Tech stack | — | — |
| Alternative to | Artificial Analysis | — |
| AI Launch upvotes | ▲ 0 | ▲ 0 |
| Hacker News | Y▲ 153 on HN | Y▲ 78 on HN |
| Launched | 2026-09-26 | 2026-09-23 |
Overview
Who it's for
JevBench
Teams choosing a decision or classification model who need to weigh accuracy against speed and cost.PacBench
Developers and AI enthusiasts comparing how well models build a working game in one shot.Problem
JevBench
Comparing models fairly is hard when providers report different benchmarks and pricing.PacBench
Abstract benchmark scores don't show what a model can actually build from a single prompt.Solution
JevBench
A reproducible benchmark with consistent scoring and cost comparisons across Jev-class models and providers.PacBench
A gallery of one-shot Pac-Man builds with the cost, time and token counts of each run.What makes it unique
JevBench
Filters for EU hosting and data-retention policy, with hosting checked per model against provider documentation.PacBench
Every entry answers the same one-line prompt, so results are directly comparable and playable.