TowerBase 13B v0.1 needs about 15.1 GB of memory at Q4_K_M (a 7.9 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 22 quants, from 2.9 GB (IQ1_S) to 10.7 GB (Q6_K).
TowerBase 13B v0.1 – a 13B multilingual LLM based on Llama 2
TowerBase 13B v0.1 is a 13B parameter language model built on Llama 2 and continued pretraining on 20 billion tokens of multilingual data. It supports English, Portuguese, Spanish, French, German, Dutch, Italian, Korean, Chinese, and Russian. The model is released under CC‑BY‑NC‑4.0.
Good at
- Translation and related tasks in supported languages
- Few‑shot and zero‑shot multilingual performance
- Fine‑tuning for APE and GEC
Limits
- No vision support
- Not aligned to human preferences
- Not guaranteed for languages outside the 10 supported
Best for
- Multilingual translation projects
- Fine‑tuning for language‑specific tasks
- Research in multilingual NLP
Who it’s for. Researchers and developers working on multilingual translation and related tasks. The model is suitable for fine‑tuning on specific language pairs or for zero‑shot inference in the supported languages.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 2.9 GB | 6.8 GB | Fits | Get file → | |
| IQ1_M | 3.1 GB | 7.0 GB | Fits | Get file → | |
| IQ2_XXS | 3.5 GB | 7.4 GB | Tight | Get file → | |
| IQ2_XS | 3.9 GB | 7.7 GB | Tight | Get file → | |
| IQ2_S | 4.2 GB | 8.1 GB | Too big | Get file → | |
| Q2_K_S | 4.4 GB | 8.3 GB | Too big | Get file → | |
| IQ2_M | 4.5 GB | 8.4 GB | Too big | Get file → | |
| Q2_K | 4.9 GB | 8.7 GB | Too big | Get file → | |
| IQ3_XXS | 5.0 GB | 8.8 GB | Too big | Get file → | |
| IQ3_XS | 5.4 GB | 9.2 GB | Too big | Get file → | |
| IQ3_S | 5.7 GB | 9.5 GB | Too big | Get file → | |
| Q3_K_S | 5.7 GB | 9.5 GB | Too big | Get file → | |
| IQ3_M | 6.0 GB | 9.8 GB | Too big | Get file → | |
| Q3_K_M | 6.3 GB | 10.2 GB | Too big | Get file → | |
| Q3_K_L | 6.9 GB | 10.8 GB | Too big | Get file → | |
| IQ4_XS | 7.0 GB | 10.8 GB | Too big | Get file → | |
| Q4_0 | 7.4 GB | 11.2 GB | Too big | Get file → | |
| Q4_K_S | 7.4 GB | 11.3 GB | Too big | Get file → | |
| Q4_K_M | 7.9 GB | 11.7 GB | Too big | Get file → | |
| Q5_K_S | 9.0 GB | 12.8 GB | Too big | Get file → | |
| Q5_K_M | 9.2 GB | 13.1 GB | Too big | Get file → | |
| Q6_K | 10.7 GB | 14.5 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (file size 2.9 GB, memory 10.1 GB at 8k context). For 12 GB VRAM, Q4_K_M (file size 7.87 GB, memory 15.1 GB) offers a good balance. For 16 GB VRAM, Q5_K_M (file size 9.23 GB, memory 16.4 GB) provides higher quality.
What GPU do you need for TowerBase 13B v0.1?
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: the best quant that fits is IQ2_M (4.5 GB file, about 11.7 GB in use).
- 16 GB VRAM: the best quant that fits is Q4_K_M (7.9 GB file, about 15.1 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (10.7 GB file, about 17.9 GB in use).
Which TowerBase 13B v0.1 quant should you choose?
Q4_K_M is the usual starting point for TowerBase 13B v0.1: it balances file size and quality well. Here it is a 7.9 GB file.
If memory is tight, the smallest file is IQ1_S at 2.9 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 TowerBase 13B v0.1
How much VRAM does TowerBase 13B v0.1 need?
About 15.1 GB at Q4_K_M with an 8K context window. That is the 7.9 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 TowerBase 13B v0.1 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of TowerBase 13B v0.1 GGUF?
It depends on the quant: IQ1_S is 2.9 GB, IQ1_M is 3.1 GB, IQ2_XXS is 3.5 GB, IQ2_XS is 3.9 GB, IQ2_S is 4.2 GB, Q2_K_S is 4.4 GB and 16 more.
What context length does TowerBase 13B v0.1 support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 6.7 GB at 8K and 3.4 GB at 4K.
Where can I download TowerBase 13B v0.1 GGUF?
The files are on Hugging Face in the mradermacher/TowerBase-13B-v0.1-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 TowerBase 13B v0.1 use?
Hugging Face lists the license as cc-by-nc-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 TowerBase 13B v0.1 on 8 GB VRAM?
Yes, the IQ1_S quant requires about 10.1 GB of VRAM at 8k context, which is close to 8 GB but may need swapping or lower context.
Does the model support vision tasks?
No, TowerBase 13B v0.1 is a purely language model and has no vision capabilities.
Is the model aligned to user preferences?
No, the model has not been aligned to human preferences and may generate problematic outputs.