Targeted Grammar and Spell Checking with LFM2.5 Encoder

This demo shows an English spelling and grammar checker built with Liquid AI's 350M-parameter LFM2.5 Encoder. The model reads the complete sentence and assigns each subword one of four edit actions: keep, delete, replace, or append. This produces targeted corrections while leaving unrelated text unchanged. Unlike open-ended text generation, the encoder performs structured editing without a token-by-token decoding loop. That makes it a practical fit for systems where precise corrections are more important than rewriting the user's text. 🔗 Links: • Hugging Face demo: https://huggingface.co/spaces/LiquidAI/spellchecker • Blog post: https://www.liquid.ai/blog/lfm2-5-encoders Connect with Liquid AI: • Careers: https://www.liquid.ai/careers • Hugging Face: https://huggingface.co/LiquidAI • Discord: https://discord.com/invite/liquid-ai • X: https://x.com/LiquidAI • LinkedIn: https://www.linkedin.com/company/liquid-ai-inc/ • GitHub: https://github.com/Liquid4All/cookbook • Substack: https://liquidai.substack.com/