madlad400 7b mt needs about 5.7 GB of memory at its main file (a 4.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 2 quants, from 4.7 GB to 6.8 GB.
MADLAD-400-7B-MT: 8.3B multilingual translation model
MADLAD-400-7B-MT is a multilingual machine translation model based on T5 architecture. It was trained on 250 billion tokens covering over 450 languages. The model is licensed under Apache 2.0.
Good at
- Translation across 400+ languages
- Large multilingual coverage
- Competitive performance with larger models
Limits
- No vision support
- No domain‑specific tuning
- Requires 8 k context for memory estimates
Best for
- Machine translation tasks
- Multilingual NLP research
- Cross‑lingual text generation
Who it’s for. Researchers and developers needing a multilingual translation model. The model is suitable for academic experiments and prototyping but is not evaluated for production use.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 4.7 GB | 5.7 GB | Fits | Get file → | |
| — | 6.8 GB | 8.1 GB | Too big | Get file → |
For a 8 GB VRAM setup, the 4.67 GB quantized file (est. 5.7 GB memory at 8 k context) is recommended. For 12 GB VRAM, the 6.81 GB file (est. 8.1 GB memory) can be used. Memory estimates are approximate.
What GPU do you need for madlad400 7b mt?
These figures add the file size, an fp16 KV cache at an 8K context window 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: the best quant that fits is the main file (4.7 GB file, about 5.7 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (4.7 GB file, about 5.7 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (6.8 GB file, about 8.1 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about madlad400 7b mt
How much VRAM does madlad400 7b mt need?
About 5.7 GB at its main file with an 8K context window. That is the 4.7 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 madlad400 7b mt on an 8 GB GPU?
Yes, with the smaller files: the 4.7 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of madlad400 7b mt GGUF?
It depends on the quant: file is 4.7 GB, file is 6.8 GB.
Where can I download madlad400 7b mt GGUF?
The files are on Hugging Face in the google/madlad400-7b-mt repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does madlad400 7b mt use?
Hugging Face lists the license as apache-2.0. 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 MADLAD-400-7B-MT on 8 GB VRAM?
Yes, using the 4.67 GB quantized file which estimates 5.7 GB memory at 8 k context.
Is MADLAD-400-7B-MT suitable for production?
It has not been assessed for production use cases and is intended for research.