What changed in AI today

• Updated 8B and 70B weights released on Hugging Face July 24, 2026, under Apache 2.0 license.
• Context window grows to 262,144 tokens, four times Apertus 1.0; models accept images and audio but output only text.
• Post-training includes Swiss Legal parity with leading open-weight models; 8B reportedly matches GPT-OSS-20B on translation.

Why it matters for marketers: Content teams can test image and audio input and longer documents on open weights, with Apache 2.0 licensing for commercial use.

• 24B total parameters with 2B active per token, built as a reasoning model that thinks before answering.
• Scores 48.09 on LiveClaw versus 38.19 for Qwen3.6-35B-A3B, a model with roughly 50% more parameters.
• Decodes 184 tokens/s on an Apple M5 Max and 114 tokens/s on a Ryzen AI Max+ 395, per the source.
• Open-weight and available on Hugging Face now; supports llama.cpp, MLX, vLLM, SGLang, and ONNX from day one.

Why it matters for marketers: Content teams building tool-using assistants can run a local agent model, but should pair it with retrieval or search for knowledge-heavy tasks.

• 1.6 billion parameter model; also transcribes English, French, Spanish and Portuguese with the same weights.
• Averages 20.92% WER across six Arabic test sets, versus 23.17% for the best published leaderboard result.
• Internal Emirati evaluation: 22.73% WER and 10.19% CER, 4.07 points below Qwen3-Omni, the next best.
• Live now as a Hugging Face Demo Space; API access and native applications are planned, not yet available.

Why it matters for marketers: Teams producing Arabic or Gulf-dialect transcripts or captions can test a free demo now, though API access is not yet released.