Apertus 1.5 Release Notes

Hugging Face
Read the full article
Apertus 1.5 models, including 8B and 70B versions, now support multimodal input with images and audio, feature a thinking mode for improved reasoning, a quadrupled context window, enhanced instruction-following, and better tool integration. The updated models were released under Apache 2.0 on July 24, 2026, with full openness on weights, data, and training details.

More in LLM & Text Generation

PyTorch Conference 2026: Google and Meta Open-Source TorchTPU for Native PyTorch on TPUs

InfoQ (AI, ML & Data)

Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model

MarkTechPost

Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster

MarkTechPost