JiRackTernaryGemma4 26b 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 6 quants, from 13.3 GB (Q3_K_M) to 50.5 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 32.2 GB.
JiRackTernaryGemma4 26b – a ternary‑quantized Gemma 4 26B model
JiRackTernaryGemma4 26b is a 26B‑parameter Gemma 4 model migrated to a ternary architecture. It removes the vision tower and keeps a text‑only decoder with 128 experts, top‑8 routing. The model supports large context windows up to 262,144 tokens and is optimized for local inference.
Good at
- Coding and tool‑calling
- Large‑context text generation
- Reasoning tasks
Limits
- No vision support
- Requires large VRAM for full precision
- License is MIT
Best for
- Local code assistants
- Long‑form document summarization
- Agentic workflows
Who it’s for. Developers and researchers who need a high‑capacity text‑only model that can run on local GPUs or CPUs with GGUF support. The ternary quantization reduces memory usage while maintaining performance.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K_M | 13.3 GB | 21.8 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 → | |
| Q6_K | 22.6 GB | 31.2 GB | Too big | Get file → | |
| Q8_0 | 26.9 GB | 35.4 GB | Too big | Get file → | |
| F16 | 50.5 GB | 59.1 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M (file size 16.8 GB, estimated 25.3 GB memory at 8 k context). For 12 GB VRAM, Q5_K_M (19.13 GB, 27.7 GB memory). For 16 GB VRAM, Q6_K (22.64 GB, 31.2 GB memory). All estimates are approximate.
What GPU do you need for JiRackTernaryGemma4 26b?
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 Q3_K_M (13.3 GB file, about 21.8 GB in use).
- 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 Q8_0 (26.9 GB file, about 35.4 GB in use).
Which JiRackTernaryGemma4 26b quant should you choose?
Q4_K_M is the usual starting point for JiRackTernaryGemma4 26b: it balances file size and quality well. Here it is a 16.8 GB file.
The Q8_0 file is 26.9 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 Q3_K_M at 13.3 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 JiRackTernaryGemma4 26b
How much VRAM does JiRackTernaryGemma4 26b 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 JiRackTernaryGemma4 26b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 21.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of JiRackTernaryGemma4 26b GGUF?
It depends on the quant: Q3_K_M is 13.3 GB, Q4_K_M is 16.8 GB, Q5_K_M is 19.1 GB, Q6_K is 22.6 GB, Q8_0 is 26.9 GB, F16 is 50.5 GB.
What context length does JiRackTernaryGemma4 26b 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 JiRackTernaryGemma4 26b GGUF?
The files are on Hugging Face in the CMSManhattan/JiRackTernaryGemma4_26b repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does JiRackTernaryGemma4 26b use?
Hugging Face lists the license as mit. 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 JiRackTernaryGemma4 26b on 8 GB VRAM?
Use the Q4_K_M quantization; it requires about 25.3 GB of VRAM at 8 k context, so 8 GB is insufficient.
Does this model support image input?
No, the vision tower has been removed for memory optimization.
What tokenizer should I use?
Use CMSManhattan/GemmaRoboticsTokenizer for best compatibility.