Meta's Omnilingual MT for 1,600 Languages • Hacker News Board

Re: Meta's Omnilingual MT for 1,600 Languages

#21

Post by ks2048 » Sat, Mar 21, 2026, 5:05 PM UTC

stingraycharles wrote: post #3

I find that meta’s translations are very poor compared to others, at least for relatively obscure languages, which I figured was relevant considering the article. Google Translate is a good default, but LLMs are really good at translations, as they’re better capable at understanding context and providing culturally appropriate translations. (I live in Cambodia where they speak Khmer)

So, LLMs are noticeably better in Khmer than Google Translate? I wonder why Google Translate doesn't use Gemini under-the-hood. Perhaps it's more prone to hallucinations.

I'm interested in find some thorough testing of translations on different LLMs vs Translation APIs.

Re: Meta's Omnilingual MT for 1,600 Languages

#22

Post by pattilupone » Sat, Mar 21, 2026, 5:10 PM UTC

ks2048 wrote: post #21

stingraycharles wrote: post #3

I find that meta’s translations are very poor compared to others, at least for relatively obscure languages, which I figured was relevant considering the article. Google Translate is a good default, but LLMs are really good at translations, as they’re better capable at understanding context and providing culturally appropriate translations. (I live in Cambodia where they speak Khmer)

So, LLMs are noticeably better in Khmer than Google Translate? I wonder why Google Translate doesn't use Gemini under-the-hood. Perhaps it's more prone to hallucinations. I'm interested in find some thorough testing of translations on different LLMs vs Translation APIs.

There's a dropdown on Google Translate that lets you choose "Advanced" mode or "Classic" mode. Advanced mode uses Gemini but it's only available for select languages.

Re: Meta's Omnilingual MT for 1,600 Languages

#23

Post by ks2048 » Sat, Mar 21, 2026, 5:10 PM UTC

I'll be looking at this in detail. I've started a company to do similar things, https://6k.ai

I'm currently concentrating on better data gathering for low-resource languages.

When you look in detail at data like Common Crawl, finepdfs, and fineweb, (1) they are really lacking quality data sources if you know where to look, and (2) the sources they have are not processed "finely" enough (e.g. finepdfs classify each page of PDF as having a specific language, where-as many language learning sources have language pairs, etc.

Re: Meta's Omnilingual MT for 1,600 Languages

#24

Post by ks2048 » Sat, Mar 21, 2026, 5:15 PM UTC

Meta released No Language Left Behind (NLLB) [1], I think in 2022. I wonder why this in not "NLLB 2.0"? These companies love introducing new names to confuse things

[1] https://ai.meta.com/research/no-language-left-behind/

Re: Meta's Omnilingual MT for 1,600 Languages

#25

Post by ks2048 » Sat, Mar 21, 2026, 5:16 PM UTC

Another interesting thing mentioned here is: BOUQuET: Benchmark and Open-initiative for Universal Quality Evaluation in Translation.

https://huggingface.co/spaces/facebook/bouquet

Re: Meta's Omnilingual MT for 1,600 Languages

#26

Post by intended » Sat, Mar 21, 2026, 5:28 PM UTC

ks2048 wrote: post #23

I'll be looking at this in detail. I've started a company to do similar things, https://6k.ai I'm currently concentrating on better data gathering for low-resource languages. When you look in detail at data like Common Crawl, finepdfs, and fineweb, (1) they are really lacking quality data sources if you know where to look, and (2) the sources they have are not processed "finely" enough (e.g. finepdfs classify each pa…

There’s many nation states working on this, have you looked into availability of those data sets?

What languages are you prioritizing?

Re: Meta's Omnilingual MT for 1,600 Languages

#27

Post by ks2048 » Sat, Mar 21, 2026, 5:29 PM UTC

djoldman wrote: post #18

Just spent a long time trying to find where you can download any of these weights. Is it open weight? If so, why isn't there just a straight link to the models?

I haven't seen anywhere claiming they are open weight (although their last similar model, NLLB was).

They say their leaderboard and evaluation datasets are freely available. Closest statement I've seen in the paper, "Our translation models are built on top of freely available models."

Re: Meta's Omnilingual MT for 1,600 Languages

#28

Post by canjobear » Sat, Mar 21, 2026, 5:33 PM UTC

garyclarke27 wrote: post #15

They can translate 1600 languages, but they cannot do basic text formatting, where are the paragraphs?

It's an abstract for a paper, so it's officially supposed to be one paragraph.

Re: Meta's Omnilingual MT for 1,600 Languages

#29

Post by ks2048 » Sat, Mar 21, 2026, 5:41 PM UTC

intended wrote: post #26

ks2048 wrote: post #23

I'll be looking at this in detail. I've started a company to do similar things, https://6k.ai I'm currently concentrating on better data gathering for low-resource languages. When you look in detail at data like Common Crawl, finepdfs, and fineweb, (1) they are really lacking quality data sources if you know where to look, and (2) the sources they have are not processed "finely" enough (e.g. finepdfs classify each pa…

There’s many nation states working on this, have you looked into availability of those data sets? What languages are you prioritizing?

Yes, there are government datasets, languge "acadamies" (or "regulators") - organizations focused on preserving / teaching the language, and often smaller, local publishers that publish material in their local language.

I'm living in Guatemala, so have been focusing on the Mayan languages here (22 languages, millions of speakers).

Re: Meta's Omnilingual MT for 1,600 Languages

#30

Post by intended » Sat, Mar 21, 2026, 5:45 PM UTC

Didn’t research show that models get worse at translation the more languages get added in? The curse of multilinguality? Lauscher 2020?

It looks like meta found a way forward.

Reading meta’s abstract, it seems that they have found ways to improve the quality of the training data, and also new evaluation tools?

They are also saying that OMT-LLaMA does a better job at text generation than other baseline models.