[KongchangAI]
BenchmarkGraphWalks

GraphWalks

A long-context test benchmark supporting million-token scales, evaluating a model's information retrieval and reasoning capabilities over long texts.

Timeline (last 90 days)

Nov 21

On long-context benchmarks, Neon scores 75.8 on Agent's Last Exam, 84 on Riemann BenchMath, and 95.2% on GraphWalks (million-token window), while Opus scores 66.8% on GraphWalks.

Unverified50%
Nov 19

In the long-context test GraphWalks, Neon scored 95.2% while Opus scored only 66.8%.

Unverified50%
Oct 18

In the GraphWalks long-context test, Argon led the compared models.

Unverified50%

All Facts (3)

Source Articles