Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

NPHardEval - GCP — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 12 models tracked.

Top models

#ModelScore
1Mistral 7B Instruct (v0.1)20.91
2GPT-3.5 Turbo8.36
3GPT-4 Turbo7.64
4vicuna-13B-v1.34.73
5Claude Instant2.91
6Claude 22.36
7Qwen-14B-Chat2.36
8Phi-1.52
9Yi 34B (Chat)1.64
10Phi-21.27
11mpt-30B Instruct0

Interactive version: theaggregate.ai/benchmark?slug=nphardeval-gcp · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.