Search public pages, research tools, and SAVRN solutions.
Open-weight model · Translation
by AI at Meta facebook/mbart-large-50-many-to-many-mmt
This model is a fine-tuned checkpoint of mBART-large-50. mbart-large-50-many-to-many-mmt is fine-tuned for multilingual machine translation. It was introduced in Multilingual Translation with Extensible Multilingual Pretraining and Finetuning paper.
What it takes to serve mbart-large-50-many-to-many-mmt (611M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 1.2 GB | 1.5 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.6 GB | 0.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.3 GB | 0.4 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
This model is a fine-tuned checkpoint of mBART-large-50. mbart-large-50-many-to-many-mmt is fine-tuned for multilingual machine translation. It was introduced in Multilingual Translation with Extensible Multilingual Pretraining and Finetuning paper. The model can translate directly between any pair of 50 languages. To translate into a target language, the target language id is forced as the first generated token. To force the target language id as the first generated token, pass the forcedbostokenid parameter to the generate method. See the model hub to look for more fine-tuned versions. Arabic (arAR), Czech (csCZ), German (deDE), English (enXX), Spanish (esXX), Estonian (etEE), Finnish…
Excerpt from the card by AI at Meta.
12 files, 12.2 GB in total. The weights are 5 files totalling 12.2 GB in bin, h5, msgpack, ot, safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| flax_model.msgpack | Weights | 2.4 GB | f38a4833c93d |
| model.safetensors | Weights | 2.4 GB | 10ef2c7b94d9 |
| pytorch_model.bin | Weights | 2.4 GB | 024ddcc796a3 |
| rust_model.ot | Weights | 2.4 GB | b4169d2f8ae7 |
| tf_model.h5 | Weights | 2.4 GB | faef970c14a4 |
| config.json | Configuration | 1.4 KB | — |
| generation_config.json | Configuration | 261 B | — |
| special_tokens_map.json | Configuration | 649 B | — |
| README.md | Documentation | 3.9 KB | — |
| sentencepiece.bpe.model | Other | 5.1 MB | cfc8146abe2a |
| .gitattributes | Repository | 794 B | — |
| tokenizer_config.json | Tokenizer | 529 B | — |
Released by AI at Meta through its official repository on Hugging Face.
| Precision | Weights in memory |
|---|---|
| As published | 12.2 GB |
| 16-bit | 1.2 GB |
| 8-bit | 0.6 GB |
| 4-bit | 0.3 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 1.5 GB at 16-bit and 0.4 GB at 4-bit: the weights (611M parameters) plus a working margin. A long context needs more.
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
1,024 tokens, from the maximum position embeddings in its published configuration.
This is mbart-large-cc25, finetuned on wmtenro. It scores BLEU 28.1 without post processing and BLEU 38 with postprocessing. Instructions in romanianpostprocessing.md Original Code: https://github.com/pytorch/fairseq/tree/master/examples/mbart Docs: https://huggingface.co/transformers/master/modeldoc/mbart.html
The developers of the Text-To-Text Transfer Transformer (T5) write: T5-Large is the checkpoint with 770 million parameters. The developers write in a blog post that the model: See the blog post and research paper for further details. The model is pre-trained on the Colossal Clean Crawled Corpus (C4), which was developed and released in the context of the same research paper as T5. The model was pre-trained on a on a multi-task mixture of unsupervised (1.) and supervised tasks (2.). Thereby, the following datasets were being used for (1.) and (2.): 1. Datasets used for Unsupervised denoising objective: 2. Datasets used for Supervised text-to-text language modeling objective - CoLA Warstadt…
This is the model card of IndicTrans2 En-Indic Distilled 200M variant. Please refer to section 7.6: Distilled Models in the TMLR submission for further details on model training, data and metrics. Please refer to the github repository for a detail description on how to use HF compatible IndicTrans2 models for inference. - New RoPE based IndicTrans2 models which are capable of handling sequence lengths upto 2048 tokens are available here - These models can be used by just changing the modelname parameter. Please read the model card of the RoPE-IT2 models for more information about the generation. - It is recommended to run these models with flashattention2 for efficient generation. If you…
Neural machine translation model for translating from English (en) to Bulgarian (bg). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. You can also use OPUS-MT models with the transformers pipelines, for example: The work is supported by the European Language Grid as pilot…
Neural machine translation model for translating from Turkish (tr) to English (en). This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. You can also use OPUS-MT models with the transformers pipelines, for example: The work is supported by the European Language Grid as pilot…
The developers of the Text-To-Text Transfer Transformer (T5) write: T5-Base is the checkpoint with 220 million parameters. The developers write in a blog post that the model: See the blog post and research paper for further details. The model is pre-trained on the Colossal Clean Crawled Corpus (C4), which was developed and released in the context of the same research paper as T5. The model was pre-trained on a on a multi-task mixture of unsupervised (1.) and supervised tasks (2.). Thereby, the following datasets were being used for (1.) and (2.): 1. Datasets used for Unsupervised denoising objective: 2. Datasets used for Supervised text-to-text language modeling objective - CoLA Warstadt et…