Technology

Multiverse Computing unveils Quasar 438B reasoning model

The company positions the new bilingual enterprise system as Europe's top-performing reasoning AI model based on benchmark testing.

Updated Sep 2, 2026 · 13:19 UTC1 reports

The short version

Multiverse Computing launched Quasar 438B, a large reasoning model designed for coding, enterprise agents, and long-context analysis.
The bilingual system operates in English and Spanish and is accessible via the company's CompactifAI API.
Company disclosures highlight high benchmark scores relative to other European models, alongside fast 500-token response times.
Independent verification across broader industry benchmarks remains pending following the vendor announcement.

Multiverse Computing launched Quasar 438B, its first large reasoning model targeted at enterprise coding and agentic workflows.[Hacker News]
Quasar 438B supports English and Spanish and is available to developers through the CompactifAI API.[Hacker News]
The model achieved a score of 43 on the Artificial Analysis Intelligence Index v4.1.1, which the company claims is the highest among European models.[Hacker News]
According to company figures, Quasar 438B generates a 500-token response, including reasoning time, in 15.3 seconds.[Hacker News]
In specific evaluations, Quasar 438B scored 75.0 on the AA-LCR long-context test and 69.3 on Terminal-Bench v2.1.[Hacker News]

Performance claims, benchmark rankings, and operational latency figures stem from vendor-provided announcements and have not been independently audited across diverse real-world workloads.[Hacker News]