AI models
Apertus 70B
Training compute
6.7×10²⁴ FLOP
Parameters
70B
Published
Sep 2, 2025
Apertus 70B is an AI model developed by ETH Zurich, Ecole Polytechnique Federale de Lausanne (EPFL), Swiss National Supercomputing Centre (CSCS) and Swisscom (Switzerland), first published in September 2025. It works in the language domain, on tasks such as language modeling/generation, question answering, code generation and translation.
Training it took an estimated 6.7×10²⁴ FLOP of compute (estimation method: reported,operation counting,hardware). The model has 70,000,000,000 parameters. It was trained on roughly 15T datapoints. Training ran on 4,096 NVIDIA GH200.
Access: Open weights (unrestricted). Its weights are openly available. Epoch AI rates the confidence of this record as confident.
Full record
- Organization
- ETH Zurich, Ecole Polytechnique Federale de Lausanne (EPFL), Swiss National Supercomputing Centre (CSCS), Swisscom
- Country of organization
- Switzerland
- Domain
- Language
- Task
- Language modeling/generation, Question answering, Code generation, Translation
- Training compute
- 6.7×10²⁴ FLOP
- Compute estimation method
- Reported, Operation counting, Hardware
- Parameters
- 70,000,000,000
- Dataset size
- 15T
- Training hardware
- NVIDIA GH200
- Chips used
- 4,096
- Chip-hours
- 6M
- Training power draw
- 5.6 MW
- Numerical format
- BF16
- Model accessibility
- Open weights (unrestricted)
- Open weights
- Yes
- Epoch confidence
- Confident
More from ETH Zurich,Ecole Polytechnique Federale de Lausanne (EPFL),Swiss National Supercomputing Centre (CSCS),Swisscom
SourceEpoch AI, 'AI Models'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/data/ai-models. Licensed under CC BY 4.0.
← All ai models