Google DeepMind · gemma4
Can I run gemma4?
Gemma 4 models are designed to deliver frontier-level performance at each size. They are well-suited for reasoning, agentic workflows, coding, and multimodal understanding.
8B parameters128K contextimagetextApache License11 builds
gemma4 system requirements
Memory needed at 4K context, counting model weights, KV cache, compute buffers and runtime overhead. Download sizes come from the official Ollama registry.
| Build | Quantization | Download | Memory needed |
|---|---|---|---|
| gemma4:e2b-it-q4_K_M | Q4_K_M | 6.7 GB | ~9.9 GB(est.) |
| gemma4:12b | Q4_K_M | 7.0 GB | ~10.2 GB(est.) |
| gemma4:e2b-it-q8_0 | Q8_0 | 7.6 GB | ~11.2 GB(est.) |
| gemma4:e4b-it-q4_K_M | Q4_K_M | 8.9 GB | ~13.2 GB(est.) |
| gemma4:e4b-it-q8_0 | Q8_0 | 10.8 GB | ~15.9 GB(est.) |
| gemma4:12b-it-q8_0 | Q8_0 | 12.0 GB | ~17.3 GB(est.) |
| gemma4:26b-a4b-it-q4_K_M | Q4_K_M | 16.8 GB | ~24.3 GB(est.) |
| gemma4:26b | Q4_K_M | 17.3 GB | ~23.0 GB(est.) |
Figures marked (est.) approximate the KV cache because this model's architecture details are not published. They are indicative rather than exact.
How to run gemma4 locally
- 1.Install Ollama for Windows, macOS or Linux.
- 2.Run this in a terminal:
ollama run gemma4:e2b-it-q4_K_M - 3.The weights download on first run, then an interactive prompt opens.
Frequently asked questions about gemma4
- How much VRAM do I need to run gemma4?
- At 4K context, the smallest published build of gemma4 (Q4_K_M) needs roughly 9.9 GB of GPU memory once weights, KV cache, compute buffers and runtime overhead are counted. Bigger quantizations and longer context windows need more. It can also run partly or entirely on the CPU using system RAM, more slowly.
- Can I run gemma4 without a dedicated GPU?
- Yes, but slowly. Without a GPU the model runs on the CPU using system RAM, which usually means a few tokens per second rather than dozens. You would need at least 9.9 GB of free RAM for the smallest build.
- How big is the gemma4 download?
- The smallest published build is 6.7 GB. There are 11 builds in total, the largest being 31.5 GB. Leave some extra free disk space beyond the download itself.
- How do I run gemma4 locally?
- Install Ollama, then run "ollama run gemma4:e2b-it-q4_K_M" in a terminal. The weights download on first use and an interactive session opens.
Will gemma4 run on your PC?
Scan your hardware once and get a verdict for this model and every build — free, no account.
Model data from the official Ollama registry · verified 13/09/2026