Gemma 4 26B A4B It Abliterated needs about 31.2 GB of memory at Q6_K (a 22.6 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 3 quants, from 19.1 GB (Q5_K_M) to 50.5 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 32.2 GB.
Gemma 4 26B A4B It Abliterated – a multimodal Mixture‑of‑Experts model
Gemma 4 26B A4B It Abliterated is a 27‑billion‑parameter model with 4‑billion active experts. It supports text, image, and tool‑calling input and outputs text. The model is available in GGUF format with Q5_K_M, Q6_K, and BF16 quantizations.
Good at
- Multimodal reasoning
- Large‑context text generation
- Tool‑calling
Limits
- Requires 256K token context window
- Vision support only
- Large VRAM for full‑precision
Best for
- Chatbots needing vision and reasoning
- Coding assistants with function calling
- Multimodal research prototypes
Who it’s for. Developers and researchers who need a powerful, multimodal LLM that can run locally with a GPU. The model is suited for edge and server deployments where a 256K context window and vision capabilities are required.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q5_K_M | 19.1 GB | 27.7 GB | Too big | Get file → | |
| Q6_K | 22.6 GB | 31.2 GB | Too big | Get file → | |
| BF16 | 50.5 GB | 59.1 GB | Too big | Get file → |
For a 16 GB GPU, use Q5_K_M (file size 19.13 GB, estimated 27.7 GB memory at 8 k context). If VRAM is limited, Q6_K (22.64 GB, 31.2 GB memory) is an alternative. BF16 offers the highest quality but needs 50.51 GB and 59.1 GB memory.
What GPU do you need for Gemma 4 26B A4B It Abliterated?
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: no quant fits at an 8K context.
- 32 GB VRAM: the best quant that fits is Q6_K (22.6 GB file, about 31.2 GB in use).
- 48 GB VRAM: the best quant that fits is Q6_K (22.6 GB file, about 31.2 GB in use).
Which Gemma 4 26B A4B It Abliterated quant should you choose?
Q6_K (22.6 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.
If memory is tight, the smallest file is Q5_K_M at 19.1 GB, with the quality trade-offs that come with fewer bits. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Gemma 4 26B A4B It Abliterated
How much VRAM does Gemma 4 26B A4B It Abliterated need?
About 31.2 GB at Q6_K with an 8K context window. That is the 22.6 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 Gemma 4 26B A4B It Abliterated on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 27.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Gemma 4 26B A4B It Abliterated GGUF?
It depends on the quant: Q5_K_M is 19.1 GB, Q6_K is 22.6 GB, BF16 is 50.5 GB.
What context length does Gemma 4 26B A4B It Abliterated 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 Gemma 4 26B A4B It Abliterated GGUF?
The files are on Hugging Face in the SevenOfNine/Gemma-4-26B-A4B-It-Abliterated-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 Gemma 4 26B A4B It Abliterated use?
Hugging Face lists the license as gemma. 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 Gemma 4 26B A4B It Abliterated on a 16 GB GPU?
Yes, with Q5_K_M quantization the model fits on a 16 GB GPU and runs at about 34.5 tokens per second on an RTX 4080 Super.
Does the model support vision input?
Yes, the model has vision capability and can process images with variable aspect ratio.
What context length does the model support?
The model supports up to 256K tokens of context.