LM Studio announces support for Google's newly released Gemma 4 12B encoder-free multimodal model

// 2h agoMODEL RELEASE

LM Studio announces support for Google's newly released Gemma 4 12B encoder-free multimodal model

LM Studio has announced immediate local support for Google's newly launched Gemma 4 12B model. Released by Google DeepMind on June 3, 2026, Gemma 4 12B is a unified, encoder-free multimodal model designed to run efficiently on consumer-grade hardware with at least 16GB of RAM or VRAM. By projecting visual and audio inputs directly into the LLM backbone rather than using separate encoders, the model dramatically reduces latency. LM Studio users can now download, run, and chat with Gemma 4 12B locally on Mac, Windows, and Linux via GGUF or MLX formats.

// ANALYSIS

Local multimodal AI is transitioning from a niche developer experiment to a mainstream desktop capability. Google's encoder-free architecture in Gemma 4 12B significantly reduces the resource overhead of vision and audio processing, making LM Studio the perfect consumer gateway for on-device agentic workflows without cloud dependencies.

* Encoder-free efficiency: Eliminating separate vision/audio projection layers reduces memory footprints and drastically lowers multimodal latency for local hardware.

* Democratizing agentic AI: The 12B parameter size fits comfortably within consumer-grade 16GB systems, bringing near-frontier intelligence directly to edge machines.

* Ecosystem speed: LM Studio’s rapid same-day support demonstrates the high agility of local inference communities compared to traditional enterprise release cycles.

// TAGS

lm-studiogemma-4gemma-4-12blocal-aimultimodalproduct-updategoogle-deepmind

DISCOVERED

2h ago

2026-06-04

PUBLISHED

3h ago

2026-06-04

RELEVANCE

8/ 10

AUTHOR

lmstudio

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

LAUNCH10m ago

ElectricSQL co-founder Sam Willis introduces Electric Agents, advocating for a state-sync architecture that moves AI agents out of isolated sandboxes and into durable, stream-based functions and logs.

ElectricSQL co-founder Sam Willis announced the architectural design of Electric Agents, a durable runtime platform built on a state-sync fabric. Moving away from the transient nature of traditional sandboxed execution, this architecture splits agent state and compute by utilizing durable event logs (Durable Streams) and typed, addressable entities. Under this model, every action, tool call, and state transition is appended to a persistent stream, enabling features such as session replaying, low-overhead stream forking, and the ability for agents to safely scale to zero and resume seamlessly.

LAUNCH16m ago

Electric SQL launches Electric Agents, challenging the VM sandbox status quo with a stream-backed, serverless architecture for durable agent workflows.

Electric SQL has launched Electric Agents, a platform designed for building durable, serverless AI agents. Instead of running agentic workflows inside expensive, stateful VM sandboxes, Electric Agents introduces a stateless model where agent logic runs inside lightweight serverless functions. Each agent is modeled as an addressable entity backed by an "Electric Stream"—a persistent, append-only event log acting as the agent's memory, inbox, and audit trail. This enables agents to scale to zero, sleep when idle, survive restarts, and run continuously in production while keeping compute stateless and cost-efficient.

LAUNCH23m ago

Higgsfield AI launches official Figma plugin

Higgsfield AI has released an official Figma plugin that integrates comprehensive AI-driven image and vector creation directly within the Figma design workspace. The tool allows designers to generate clean SVGs, remove backgrounds, apply color grades, swap faces while keeping scenes consistent, shoot studio product photos, and animate hero creatives, thereby eliminating the need to context-switch between Figma and external creative apps.