Atlas

Benchmarks

← All benchmarks

GDPx: Chip Design

Professional Work · 2026-11-01

GDPx's Chip Design domain: 34 tasks from Aria, Arcverex and Ergodex on verification, RTL design under power, performance and area pressure, and physical design, seeded from real taped-out silicon. Each partner's pass@1 over three attempts per task is averaged, partners weighted equally. A component of the GDPx score, which weights its 11 domains equally.

Top models (higher is better)

ModelScore
GPT-6.1 Sol5.6
Claude Opus 5.54.8
GPT-6 Astra4.4
Claude Sonnet 5.53.5
GPT-6.1 Terra2.8
Gemini 4 Argon1.6
Gemini 4 Carbon1.4
GPT-6 Luna1.3
Claude Fable 5.10.8
Claude Haiku 5.50.7
Muse Wonder Full0.6
MiniMax M3.1 Flash0.1
Muse Wonder0.1
Kimi K30.1
GLM-5.30.1
Grok 4.70.0
Muse Spark Open0.0
Loading Atlas data…