Skip to main content
← Back to news

Ollama v0.33.3: Gemma4 supports images and audio on MLX

On-device & edge
Source:Ollama
Gemma4 can now process images and audio on the MLX engine. Also, cached prompt tokens are now reported correctly and default parameters defined in GGUF files are honored.
Read original article

Related articles

Mac Studio M5 Ultra: practical 512GB guide - models that fit and GGUF setup

The 512GB Mac Studio holds a 560B MoE at 4 bit with a large context. This guide covers the 4-bit rule, GGUF setup in three paths, the real bottleneck (the KV cache), and the Mac-vs-API cost comparison with the numbers on the table.

Local Image Models: from SDXL to Qwen-Image

From SDXL's 2.6B parameters to Qwen-Image 2.1's 7B: how diffusion models work, how much VRAM you actually need, and which model to pick for your hardware.

GGUF: The Format That Keeps Models on Your Hardware

We open the box: what's really inside a .gguf file, why it's a single file, how to read the filename, and how to pick the size for your VRAM, with the real sizes of an 8B model.

Get the next local-AI briefing

Once a week, the top selection from iaLocale: articles, news and resources on local AI, and more.

After signing up you receive a confirmation email (double opt-in). Without confirmation we send nothing. No spam, we promise.

Related resources

Tools and resources from the iaLocale archive relevant to this content.

🎵

Stable Audio Open 1.0

Models

Up to 47 seconds of text-to-audio: high-quality music and sound effects. Excellent for local audio production and sound design.

Bespoke Nimble 9B

Models

Bespoke Labs' 9B decision model: it runs entirely locally via Ollama 0.35 and answers typed questions (true/false, numeric scale, multiple choice) through the new systemone API. Built for fast classification in automated pipelines, no cloud.

⚡

faster-whisper

Inference

Reimplementation of Whisper based on CTranslate2: up to 4 times faster than openai/whisper with the same accuracy and less memory. Supports 8-bit quantization on both CPU and GPU.

🎨

FLUX 3 Image Klein

Models

A 5.6B open-weights image model for text-to-image, single-reference editing and multi-reference composition in one model, with placement boxes in the prompt. Qwen3-VL-4B text encoder. In BF16 (11.1GB), FP8 and NVFP4 (5.3GB). FLUX Kommunity license (non-commercial + qualified-user exception).

Browse all resources