v0.33.3: gemma4: image and audio input support
Article image or reusable cover for Ollama
Safetensors gemma4 imports served by the MLX engine now answer image and audio chats.
Images run through both vision architectures: the transformer tower (26B, 31B, e-series) and the 12B's encoder-free unified embedder. Audio arrives through the same intake the ollama API already accepts for gemma4 GGUFs — WAV bytes in the images field, OpenAI input_audio parts, and /v1/audio/transcriptions uploads — with the e2b/e4b checkpoints running clips through their conformer audio encoder and the 12b unified checkpoint embedding the raw waveform directly. Clips longer than 30 seconds are split evenly into chunks of at most 30 seconds, cut at pauses, and encoded independently. Each modality serves only checkpoints that carry it: 26B/31B have no audio config and reject audio input, and checkpoints with an unrecognized vision architecture still load as text-only models and reject image requests. The server previously hid the vision and audio capabilities for gemma4 safetensors because the engine served neither. Both suppressions are removed, and existing imports start advertising the capabilities without re-importing since import already records them.
The text is the source's own description of its publication. The content belongs to Ollama.
More to read
Lumaril, which develops AI software for optimizing industrial production by connecting to existing control systems, raised a $66M Series A at a $400M valuation
Techmeme 4 h ago
DeepSeek open sources Super-Agent as a research prototype, a proactive coding agent it says can monitor group chats for coding issues and suggest solutions
Techmeme 5 h ago
Insurance software maker Guidewire agrees to acquire Hyperexponential, which makes AI-powered insurance underwriting and pricing software, for $1.7B in cash
Techmeme 5 h ago
One word for AI agents in 2027
Nate Herk 9 h ago