Rumeni 26B A4B needs about 25.8 GB of memory at Q4_K_M (a 17.2 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 11 quants, from 10.8 GB (Q2_K) to 27.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 32.2 GB.
26B Russian roleplay model with Mixture of Experts
Rumeni 26B A4B is a 26‑billion parameter gemma4 model fine‑tuned for Russian roleplay. It includes a LoRA on attention and dense MLP, trained on a Russian RP dataset. The model supports reasoning, roleplay, and tool use.
Good at
- Russian roleplay
- Reasoning in Russian
- Tool integration
Limits
- No vision support
- Requires large VRAM for higher quants
- License is gemma
Best for
- Chatbots in Russian
- Creative writing in Russian
- Interactive storytelling
Who it’s for. Developers building Russian language chat applications or content creators needing consistent character voices. The model is suitable for users who can run GGUF files with llama.cpp.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 10.8 GB | 19.4 GB | Too big | Get file → | |
| Q3_K_S | 12.5 GB | 21.1 GB | Too big | Get file → | |
| Q3_K_M | 13.6 GB | 22.2 GB | Too big | Get file → | |
| Q3_K_L | 14.1 GB | 22.7 GB | Too big | Get file → | |
| IQ4_XS | 14.5 GB | 23.0 GB | Too big | Get file → | |
| Q4_K_S | 15.9 GB | 24.4 GB | Too big | Get file → | |
| Q4_K_M | 17.2 GB | 25.8 GB | Too big | Get file → | |
| Q5_K_S | 18.5 GB | 27.0 GB | Too big | Get file → | |
| Q5_K_M | 19.6 GB | 28.2 GB | Too big | Get file → | |
| Q6_K | 23.2 GB | 31.8 GB | Too big | Get file → | |
| Q8_0 | 27.6 GB | 36.2 GB | Too big | Get file → |
For 8 GB VRAM use Q2_K (10.82 GB file, ~19.4 GB memory). For 12 GB VRAM use Q3_K_S (12.54 GB file, ~21.1 GB memory). For 16 GB VRAM use Q4_K_S (15.88 GB file, ~24.4 GB memory).
What GPU do you need for Rumeni 26B A4B?
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 (14.5 GB file, about 23.0 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (23.2 GB file, about 31.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (27.6 GB file, about 36.2 GB in use).
Which Rumeni 26B A4B quant should you choose?
Q4_K_M is the usual starting point for Rumeni 26B A4B: it balances file size and quality well. Here it is a 17.2 GB file.
The Q8_0 file is 27.6 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 10.8 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Rumeni 26B A4B
How much VRAM does Rumeni 26B A4B need?
About 25.8 GB at Q4_K_M with an 8K context window. That is the 17.2 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 Rumeni 26B A4B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 19.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Rumeni 26B A4B GGUF?
It depends on the quant: Q2_K is 10.8 GB, Q3_K_S is 12.5 GB, Q3_K_M is 13.6 GB, Q3_K_L is 14.1 GB, IQ4_XS is 14.5 GB, Q4_K_S is 15.9 GB and 5 more.
What context length does Rumeni 26B A4B 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 Rumeni 26B A4B GGUF?
The files are on Hugging Face in the mradermacher/Rumeni-26B-A4B-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 Rumeni 26B A4B 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 this model on 8 GB VRAM?
Use the Q2_K quant, which requires about 19.4 GB of VRAM at 8 k context.
Does the model support English?
It performs best in Russian; English performance is limited and may contain errors.
Is there a vision capability?
No, the model does not support vision.