Hugging Face Blog·· 23 h ago
Apertus 1.5 Release Notes
Apertus 1.5 Release Notes
AI summary
Our latest release, Apertus 1.5, extends the 8B and 70B Apertus 1.0 models with image/audio understanding, an optional thinking mode, a four times longer context window, improved instruction-following, and better tool use. The updated 8B and 70B model weights were released to Hugging Face on July 24, 2026 under an Apache 2.0 license.
Selection record
Threshold 60Official, first-handFirst 74Second 62
AdmittedSum of both 136 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：Apertus 1.5开源大模型发布，含多模态与训练细节
- Why it was chosen
- Apertus 1.5 adds image and audio understanding plus a thinking mode to the 8B and 70B models, and releases its training data and intermediate checkpoints, so the full training path of an open multimodal model is visible.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Hugging Face Blog · huggingface.co