{"slug": "mostik-links-models-in-latent-space", "title": "Mostik Links Models in Latent Space", "summary": "Mostik CEO Sasha Malysheva announced a new approach for model communications that links models directly in latent space, bypassing text-based interaction. In an experiment, linking GLM-5.2 with Qwen-3.5 produced results between the two models at about 1/20 the cost of running the large model alone, suggesting future AI systems could be networks of models sharing internal state.", "body_md": "# Mostik Links Models in Latent Space\n\n[Sasha Malysheva](https://x.com/aimalysheva/status/2095232794792255848), Mostik’s CEO, announcing a new approach for model communications.\n\nThey connect a large model and a small model directly through their internal representations, instead of making them communicate with text.\n\nIn one experiment, they linked GLM-5.2 with Qwen-3.5. The result performed between the two models, at about 1/20 the cost of running the large model alone.\n\nThe idea is simple: let the small model handle most of the work, and use the large model only for the hardest reasoning.\n\nIf this works, future AI systems may not be one giant model, but networks of models that share internal state directly.", "url": "https://wpnews.pro/news/mostik-links-models-in-latent-space", "canonical_source": "https://julin.ai/2026/09/03/mostik-latent-bridge/", "published_at": "2026-09-02 12:00:00+00:00", "updated_at": "2026-09-02 23:24:53.435173+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-research"], "entities": ["Mostik", "Sasha Malysheva", "GLM-5.2", "Qwen-3.5"], "alternates": {"html": "https://wpnews.pro/news/mostik-links-models-in-latent-space", "markdown": "https://wpnews.pro/news/mostik-links-models-in-latent-space.md", "text": "https://wpnews.pro/news/mostik-links-models-in-latent-space.txt", "jsonld": "https://wpnews.pro/news/mostik-links-models-in-latent-space.jsonld"}}