AION
Open-source releaseMultimodal Models · Large Language Models · MLOps, Tooling & Infrastructure1 source · Sep 29, 2026

ollama/ollama v0.35.1

Clef (27B) and Clef Flash (9B) are multimodal: requests can now include images alongside the text state, shared by all questions and scored jointly with it.

llama.cppOllamaMLX
← Back to brief

Key points

  • "state": "The user took this screenshot.",
  • "hasollama": {"type": "noul", "instructions": "Does this image contain Ollama?"}

Sources (1)

  • [1]ollama/ollama v0.35.1
    GitHub: ollama/ollama · Sep 29, 08:14 PM
    Clef (27B) and Clef Flash (9B) are multimodal: requests can now include images alongside the text state, shared by all questions and scored jointly with it.
    "state": "The user took this screenshot.",

Extractive summary: sentences quoted from the sources.