Gemma4 needs about 25.3 GB of memory at Q4_K_M (a 16.8 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 4 quants, from 11.3 GB (IQ3_XXS) to 19.1 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 32.2 GB.
Gemma4 26B A4B: 25.2B parameters, 256K context, MoE, coding and roleplay
Gemma4 26B A4B is a 25.2B parameter model with a 262144 token context window. It is a mixture of experts with 4.0B active parameters. The model supports coding and roleplay tasks and is released under Apache 2.0.
Good at
- Large context handling
- Mixture‑of‑Experts efficiency
- Coding and roleplay
Limits
- No vision or audio support in this variant
- Requires high VRAM for larger quants
- License is Apache 2.0
Best for
- Long‑form text generation
- Complex coding tasks
- Agentic roleplay scenarios
Who it’s for. Developers and researchers needing a large‑context LLM with MoE for coding or roleplay. Suitable for server‑side deployment where GPU memory allows the larger quantizations.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ3_XXS | 11.3 GB | 19.9 GB | Too big | Get file → | |
| IQ4_XS | 13.9 GB | 22.5 GB | Too big | Get file → | |
| Q4_K_M | 16.8 GB | 25.3 GB | Too big | Get file → | |
| Q5_K_M | 19.1 GB | 27.7 GB | Too big | Get file → |
For 8 GB VRAM, use IQ3_XXS (file size 11.33 GB, ~19.9 GB memory at 8 k context). For 12 GB VRAM, IQ4_XS (13.92 GB, ~22.5 GB). For 16 GB VRAM, Q4_K_M (16.8 GB, ~25.3 GB). For 20 GB VRAM, Q5_K_M (19.13 GB, ~27.7 GB).
What GPU do you need for Gemma4?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is IQ4_XS (13.9 GB file, about 22.5 GB in use).
- 32 GB VRAM: the best quant that fits is Q5_K_M (19.1 GB file, about 27.7 GB in use).
Which Gemma4 quant should you choose?
Q4_K_M is the usual starting point for Gemma4: it balances file size and quality well. Here it is a 16.8 GB file.
If memory is tight, the smallest file is IQ3_XXS at 11.3 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Gemma4
How much VRAM does Gemma4 need?
About 25.3 GB at Q4_K_M with an 8K context window. That is the 16.8 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Gemma4 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 19.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Gemma4 GGUF?
It depends on the quant: IQ3_XXS is 11.3 GB, IQ4_XS is 13.9 GB, Q4_K_M is 16.8 GB, Q5_K_M is 19.1 GB.
What context length does Gemma4 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 8.1 GB at 8K and 32.2 GB at 32K.
Where can I download Gemma4 GGUF?
The files are on Hugging Face in the wwlwwl/gemma-4-26B-A4B-it-heretic-zh-RP-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Gemma4 use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Gemma4 26B A4B on a GPU with 8 GB VRAM?
Use the IQ3_XXS quantization; it requires about 19.9 GB memory at 8 k context, so 8 GB VRAM is insufficient.
Does this model support image input?
No, this variant does not include vision or audio support.
What license applies to this model?
The model is released under the Apache 2.0 license.