Skip to main content
Evaluation sources/SWE-Atlas Codebase QnA
ObservingCoding

SWE-Atlas Codebase QnA

Scale AI · Public model evaluation results from SWE-Atlas Codebase QnA.

Official evaluation
Source description

This source measures concrete delivery performance in Coding; raw scores are shown separately from the normalized capability score.

Limitation: Different versions, reasoning settings, and test sets may change scores; read them with the data date.

How it is used

Participating sources affect capability scores; reference, cross-reference, and observing sources are for comparison and verification only.

Model results from this source

Claude Opus 4.78181.02026-08-05
GPT-5.580.880.82026-08-05
GPT-5.579.179.12026-08-05
Claude Opus 4.778.478.42026-08-05
GPT-5.57575.02026-08-05
GLM-5.173.273.22026-08-05
GPT-5.472.972.92026-08-05
GPT-5.472.472.42026-08-05
Claude Opus 4.671.971.92026-08-05
Claude Opus 4.771.771.72026-08-05
Claude Sonnet 4.670.370.32026-08-05
DeepSeek V4 Pro67.867.82026-08-05
Kimi K2.659.859.82026-08-05
Gemini 3.1 Pro Preview45.645.62026-08-05
GPT-5.545.4345.42026-08-05
GPT-5.440.840.82026-08-05
GPT-5.436.336.32026-08-05
Claude Opus 4.633.333.32026-08-05
Claude Sonnet 4.631.231.22026-08-05
Claude Opus 4.63030.02026-08-05
GLM-520.520.52026-08-05
Gemini 3.1 Pro Preview13.513.52026-08-05
Kimi K2.513.113.12026-08-05
MiniMax-M2.510.310.32026-08-05
Gemini 3 Flash Preview8.28.22026-08-05

Frequently asked questions