3 ms·
A problem I have with the open source models is that they are all not remotely good in many languages other than English compared to the OpenAI models. I specif
by jurmous 3y ago
A problem I have with the open source models is that they are all not remotely good in many languages other than English compared to the OpenAI models. I specifically need Dutch and the outputs are unusable for us.
- danielbln 3y agohttps://openchat.team/ https://openchat.team/ is quite good at German, how does it fare with Dutch?
- jug 3y agoYeah, this can be an issue. Today, there are few specialized LLM's of high quality so you end up having to use a massive all-in-one model like GPT-4 to reach for the language you need. There's movement here though and it will get better. GPT-SW3 is a new model developed by AI Sweden, trained specifically on the Nordic languages only + English. And beyond this, you have TrustLLM which is a new project that aims to be a large, open, European model trained on the Germanic languages to start with: https://liu.se/en/research/trustllm https://liu.se/en/research/trustllm
- jurmous 3y agoYes there is some movement. Good to see trustllm being another initiative. The Dutch government starts with some non profit agencies to train a Dutch model too. But this will also take some time. https://www.tno.nl/en/newsroom/2023/11/netherlands-starts-realisation-gpt-nl/ https://www.tno.nl/en/newsroom/2023/11/netherlands-starts-re...