TildeOpen 30b needs about 21.1 GB of memory at Q4_K_M (a 18.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 24 quants, from 0.0 GB to 25.2 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.
TildeOpen 30b – a 30.7B parameter multilingual decoder-only transformer
TildeOpen 30b is a 30.7B parameter dense decoder-only transformer trained on 65,536 context tokens. It supports 39 European languages and additional domains such as code and XML. The model is released under CC‑BY‑4.0 and is not yet instruction‑tuned or safety‑aligned.
Good at
- Multilingual text generation across 39 European languages
- Large context handling up to 65,536 tokens
- Open‑source foundation for future specialized models
Limits
- No instruction tuning or safety alignment
- No vision support
- Requires large VRAM for higher‑quality quants
Best for
- Research on underrepresented European languages
- Building multilingual translation systems
- Developing domain‑specific language models
Who it’s for. Researchers and developers needing a large, multilingual foundation model for European languages. The model is suitable for building specialized translation or domain models on top of the base.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 2.5 GB | Fits | Get file → | |
| IQ1_S | 6.9 GB | 9.4 GB | Too big | Get file → | |
| IQ1_M | 7.5 GB | 10.0 GB | Too big | Get file → | |
| IQ2_XXS | 8.5 GB | 11.0 GB | Too big | Get file → | |
| IQ2_XS | 9.4 GB | 11.9 GB | Too big | Get file → | |
| IQ2_S | 9.9 GB | 12.4 GB | Too big | Get file → | |
| IQ2_M | 10.7 GB | 13.2 GB | Too big | Get file → | |
| Q2_K_S | 10.8 GB | 13.3 GB | Too big | Get file → | |
| Q2_K | 11.6 GB | 14.1 GB | Too big | Get file → | |
| IQ3_XXS | 12.0 GB | 14.6 GB | Too big | Get file → | |
| IQ3_XS | 12.9 GB | 15.4 GB | Too big | Get file → | |
| Q3_K_S | 13.5 GB | 16.0 GB | Too big | Get file → | |
| IQ3_S | 13.6 GB | 16.1 GB | Too big | Get file → | |
| IQ3_M | 14.0 GB | 16.5 GB | Too big | Get file → | |
| Q3_K_M | 15.0 GB | 17.5 GB | Too big | Get file → | |
| Q3_K_L | 16.2 GB | 18.7 GB | Too big | Get file → | |
| IQ4_XS | 16.6 GB | 19.1 GB | Too big | Get file → | |
| Q4_0 | 17.5 GB | 20.0 GB | Too big | Get file → | |
| Q4_K_S | 17.6 GB | 20.1 GB | Too big | Get file → | |
| Q4_K_M | 18.5 GB | 21.1 GB | Too big | Get file → | |
| Q4_1 | 19.3 GB | 21.9 GB | Too big | Get file → | |
| Q5_K_S | 21.2 GB | 23.7 GB | Too big | Get file → | |
| Q5_K_M | 21.8 GB | 24.3 GB | Too big | Get file → | |
| Q6_K | 25.2 GB | 27.7 GB | Too big | Get file → |
For 8 GB VRAM, use the IQ1_S quant (6.9 GB file, ~9.4 GB memory at 8 k context). For 12 GB VRAM, Q4_K_M (18.54 GB file, ~21.1 GB memory) is a good balance of size and quality. Larger VRAM allows use of Q5_K_M or Q6_K for higher fidelity.
What GPU do you need for TildeOpen 30b?
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: the best quant that fits is the main file (0.0 GB file, about 2.5 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.5 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.5 GB in use).
- 12 GB VRAM: the best quant that fits is IQ2_XS (9.4 GB file, about 11.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q3_K_S (13.5 GB file, about 16.0 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_S (21.2 GB file, about 23.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (25.2 GB file, about 27.7 GB in use).
Which TildeOpen 30b quant should you choose?
Q4_K_M is the usual starting point for TildeOpen 30b: it balances file size and quality well. Here it is a 18.5 GB file.
If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about TildeOpen 30b
How much VRAM does TildeOpen 30b need?
About 21.1 GB at Q4_K_M with an 8K context window. That is the 18.5 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 TildeOpen 30b on an 8 GB GPU?
Yes, with the smaller files: the 0.0 GB file needs about 2.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of TildeOpen 30b GGUF?
It depends on the quant: file is 0.0 GB, IQ1_S is 6.9 GB, IQ1_M is 7.5 GB, IQ2_XXS is 8.5 GB, IQ2_XS is 9.4 GB, IQ2_S is 9.9 GB and 18 more.
What context length does TildeOpen 30b support?
The GGUF header lists a trained context length of 65,536 tokens (64K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.
Where can I download TildeOpen 30b GGUF?
The files are on Hugging Face in the mradermacher/TildeOpen-30b-i1-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 TildeOpen 30b use?
Hugging Face lists the license as cc-by-4.0. 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 TildeOpen 30b on 8 GB VRAM?
Use the IQ1_S quant; it requires about 9.4 GB memory at 8 k context, which is close to 8 GB VRAM but may need swapping.
Does TildeOpen 30b support image input?
No, the model is a text‑only decoder‑only transformer with no vision capabilities.
Is the model aligned with safety guidelines?
No, the base model is not instruction‑tuned or safety‑aligned; future versions will address this.