gemma
12B parameters
Commercial OK
Multimodal
Reviewed July 2026

Gemma 4 12B

Gemma 4 12B is Google's dense 12B release from April 2026, with the unified 12B variant landing in June 2026. It runs a 256K context, takes vision and audio input through an encoder-free multimodal design, and supports tool calling and thinking under an Apache-2.0 license. The gemma4 family sits at 18.7M Ollama pulls (#11 in the library); the Q4_K_M build is 7.6GB.

License: Apache-2.0·Released Apr 2, 2026·Context: 262,144 tokens

Overview

Gemma 4 12B is Google's dense 12B release from April 2026, with the unified 12B variant landing in June 2026. It runs a 256K context, takes vision and audio input through an encoder-free multimodal design, and supports tool calling and thinking under an Apache-2.0 license. The gemma4 family sits at 18.7M Ollama pulls (#11 in the library); the Q4_K_M build is 7.6GB.

Strengths

  • Apache-2.0 license with commercial use allowed
  • Encoder-free vision and audio input plus tools and thinking in one 12B build
  • 256K context at a 7.6GB Q4_K_M footprint

Weaknesses

  • 12B dense decodes slower than the 8-9B class on the same GPU
  • Filling the 256K window pushes KV cache well past what 12GB cards hold
  • The unified multimodal variant is recent (June 2026) with limited third-party quality data

Quantization variants

Each quantization trades model quality for file size and VRAM. Q4_K_M is the most popular starting point.

QuantizationFile sizeVRAM required
Q4_K_M7.6 GB10 GB

Get the model

Ollama

One-line install

ollama run gemma4:12bRead our Ollama review →

Hardware that runs this

Cards with enough VRAM for at least one quantization of Gemma 4 12B.

NVIDIA GB200 NVL72
13824GB · nvidia
AMD Instinct MI350X
288GB · amd
NVIDIA B300 (Blackwell Ultra)
288GB · nvidia
AMD Instinct MI355X
288GB · amd
AMD Instinct MI325X
256GB · amd
AMD Instinct MI300X
192GB · amd
NVIDIA B200
192GB · nvidia
NVIDIA H100 NVL
188GB · nvidia
Compare alternatives

Models worth comparing

Same parameter band, plus what's one tier above and below — so you can decide what actually fits your hardware.

Frequently asked

What's the minimum VRAM to run Gemma 4 12B?

10GB of VRAM is enough to run Gemma 4 12B at the Q4_K_M quantization (file size 7.6 GB). Higher-quality quantizations need more.

Can I use Gemma 4 12B commercially?

Yes — Gemma 4 12B ships under the Apache-2.0, which permits commercial use. Always read the license text before deployment.

What's the context length of Gemma 4 12B?

Gemma 4 12B supports a context window of 262,144 tokens (about 262K).

How do I install Gemma 4 12B with Ollama?

Run `ollama pull gemma4:12b` to download, then `ollama run gemma4:12b` to start a chat session. The default quantization is Q4_K_M.

Does Gemma 4 12B support images?

Yes — Gemma 4 12B is multimodal and accepts text + vision + audio inputs. Vision support requires a runner that handles its image-conditioning architecture.

Source: Vendor official documentation

Reviewed by RunLocalAI Editorial. See our editorial policy for how we research and verify model claims.

Related — keep moving

Compare hardware
Buyer guides
When it doesn't work
Recommended hardware
Before you buy

Verify Gemma 4 12B runs on your specific hardware before committing money.

Will it run on my hardware? →Custom hardware comparison →GPU recommender (4 questions) →