Model record
Gemma 4 12B Unified
Google DeepMind · Apache 2.0 · verified
What this model is
The architecture is dense. Every parameter is part of the active path, so memory and speed scale with the full model rather than a smaller expert subset.
Memory and quantization
Decimal GB. Q4_0 uses Google's QAT GGUF; other sizes use the published Unsloth GGUF. All totals include the multimodal projector when published separately.
| Quantization | Published files | 8K estimate | Smallest comfortable GPU | Smallest comfortable Mac | Context at smallest GPU / Mac | Speed estimate | Run commands |
|---|---|---|---|---|---|---|---|
| Q4_0 | 7.15 GB | 9.05 GB | NVIDIA GeForce RTX 3060 12GB · 12 GB VRAM | Apple Mac mini M4 16GB · 16 GB unified memory | 68,127 / 149,498 tokens | Estimate · Moderate relative class | ollama run gemma4:12b-it-qat lms get "Gemma 4 12B Unified"@q4_0 |
| Q4_K_M | 7.30 GB | 9.20 GB | NVIDIA GeForce RTX 3060 12GB · 12 GB VRAM | Apple Mac mini M4 16GB · 16 GB unified memory | 65,157 / 146,528 tokens | Estimate · Moderate relative class | ollama run gemma4:12b-it-q4_K_M lms get "Gemma 4 12B Unified"@q4_k_m |
| Q8_0 | 12.85 GB | 14.75 GB | NVIDIA GeForce RTX 4090 24GB · 24 GB VRAM | Apple Mac mini M4 24GB · 24 GB unified memory | 196,408 / 196,408 tokens | Estimate · Moderate relative class | ollama run gemma4:12b-it-q8_0 lms get "Gemma 4 12B Unified"@q8_0 |
Estimate formula: published weight files + 0.4027 GB per 8K tokens × context / 8K + 1.5 GB runtime overhead. Sizes are decimal GB, not GiB. At 8K context, the smallest listed build is about 9.1 GB before OS and application reserve.
Conservative FP16 estimate from the public config. Eight of 48 layers use global attention and local layers use a 1024-token sliding window.
Ollama tags were checked against the linked Ollama library pages on 2026-11-02. LM Studio's lms get suffix selects a file when the search results expose that exact quantization; confirm the local key with lms ls.
KV method source: https://huggingface.co/google/gemma-4-12B-it/raw/main/config.json
Speed heuristic
Not benchmarked here. The calculator labels relative speed from the dense parameter count only.
Active path: 11.95B parameters. This is a relative speed signal only; it is not a tokens/sec benchmark.
Runtimes
- Ollama
- llama.cpp
- MLX
ollama run gemma4:12blms get "Gemma 4 12B Unified"@q4_0
lms load <model-key-from-lms-ls> --gpu max --context-length 8192License restrictions
Apache 2.0; the separate Gemma Prohibited Use Policy and intended-use statement still apply.
This is a plain-language summary, not legal advice. Read the linked license before commercial deployment or redistribution.
Known issues
- The model is multimodal and the projector adds to the download and memory footprint.
- The 256K context ceiling is not a practical default on 12-16 GB machines.
If it does not fit, try these smaller records
These alternatives have a lower recommended memory target in the current catalog. Open each page to compare the quantization-specific fit before downloading.
Phi-4 Mini
Dense · starting at 5 GB
Llama 3.2 3B
Dense · starting at 4 GB
FAQ
Model-specific questions
What is the smallest local file for Gemma 4 12B Unified?
The smallest published weight package in this record is Q4_0 at 7.15 GB. The 8K runtime estimate adds 0.40 GB of KV cache and 1.5 GB overhead, for about 9.1 GB before operating-system reserve.
How much memory do the published quantizations need at 8K?
Q4_0 about 9.1 GB at 8K; Q4_K_M about 9.2 GB at 8K; Q8_0 about 14.7 GB at 8K. These are calculated estimates using the published weight sizes and the model's KV-cache assumption, not live memory measurements.
Which GPU and Mac configurations are comfortable for Gemma 4 12B Unified?
Against the current catalog, the smallest configurations that clear the 20% headroom target are: Q4_0: NVIDIA GeForce RTX 3060 12GB; Q4_K_M: NVIDIA GeForce RTX 3060 12GB; Q8_0: NVIDIA GeForce RTX 4090 24GB. A larger configuration may still be preferable for longer context or faster memory bandwidth.
What is the right Ollama or LM Studio command for Gemma 4 12B Unified?
Use the command attached to the specific quantization in the table. Ollama tags are taken from the checked Ollama library pages. For LM Studio, use lms get with the displayed quantization suffix, then load the resulting local key from lms ls with --gpu max and the desired context length.
What should I run instead if Gemma 4 12B Unified does not fit?
Smaller catalog alternatives include Phi-4 Mini, Llama 3.2 3B. Compare their quantization tables and minimum memory before choosing; a smaller parameter count does not always mean a proportionally smaller download.
Sources
- https://huggingface.co/google/gemma-4-12B-it
- https://huggingface.co/google/gemma-4-12B-it-qat-q4_0-gguf
- https://huggingface.co/unsloth/gemma-4-12b-it-GGUF
- https://ollama.com/library/gemma4
- https://huggingface.co/google/gemma-4-12B-it/raw/main/config.json
This page is an independent compatibility guide, not official documentation. Verify the primary source before downloading or deploying a model.