Multiverse Computing

Spanish AI company serving compressed and open-weight models through its CompactifAI API.

Multiverse Computing is a company from San Sebastian, Spain, that compresses large language models with quantum-inspired tensor networks. Its CompactifAI API serves its own compressed models, such as Quasar and Carina, alongside open-weight models like GLM and Qwen, on its own GPUs in Spain and the United States. A fit when you want smaller, cheaper versions of large open models.

Multiverse Computing's EU routes are also listed with every other model that runs in Europe on Opper. All EU-hosted models

2 routes8 modelsEU, US🇪🇸 HQ Spain
multiversecomputing.com

Call the 8 models on Multiverse Computing through Opper with one API key. Sign up without a credit card.

Models on Multiverse Computing

Every model we route through Multiverse Computing. Compare residency, zero data retention, training posture and price at a glance, with full data-handling detail per route below.

ModelContextRegionZero data retentionTrainingInputOutputCache read
524KEUNo$0.20$0.64$0.02
524KUSNo$0.20$0.64$0.02
262KEUNo$0.09$0.44$0.02
262KUSNo$0.09$0.44$0.02
262KUSNo$0.17$0.54$0.02
203KEUNo$0.11$0.33$0.0099
131KEUNo$0.02$0.09Input price
131KUSNo$0.02$0.09Input price

Data handling per route

Multiverse Computing hosts on 2 routes. Each route has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

Spain🇪🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. EU.

Zero data retention
Yes. Nothing is logged, held for abuse monitoring or used for training.
Training
No training on customer data.
Logging
None
Abuse monitoring
No classifier
Caching
Content cached for replay
Transfer mechanism
Not applicable — data stays in EU

United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; SCCs.

Zero data retention
Yes. Nothing is logged, held for abuse monitoring or used for training.
Training
No training on customer data.
Logging
None
Abuse monitoring
No classifier
Caching
Content cached for replay
Transfer mechanism
SCCs

Start building with 700+ models

One API key for every major provider, up and running in minutes.

Get startedView Documentation