I’m trying to use the groq moonshotai/kimi-k2-instruct model. According to the documentation on moonshotai/Kimi-K2-Instruct · Hugging Face this model isnt’ compatible with OpenAI or Gemini tokenizers and it appears to use it’s own custom tokenizer. Is it possible to configure discourse to use a cus…

TL;DR pick the closest tokenizer and set the maximum context to some thousands less to allow for the difference to not affect you.

Thanks. So I decided to enroll the services of ChatGPT, Gemini and Grok to help me decided which tokenizer to use; which would be the closest match to the Kimi Instruct TikToken/BPE tokenizer to generate the most accurate output from the model. I must say modern AI models are fairly representative …

Also worth noting … majority of coding agents these days don’t even bother with an accurate tokenizer like Discourse does. They just estimate at 4 letters per token. cl100k will be plenty fine for the vast majority of use cases on llms with slightly different tokenizers.

Como usar modelos de IA com tokenizadores customizados

Suporte

Falco (Falco) Dezembro 14, 2025, 12:02am 2

TL;DR escolha o tokenizador mais próximo e defina o contexto máximo para alguns milhares a menos para permitir que a diferença não o afete.

Tópico		Respostas	Visualizações
Adding Semantic Search feature for our self-hosted discourse site Support ai , ai-search	9	253	19 de Março de 2025
Frustrations on AI spam detector Support spam , ai	9	137	21 de Dezembro de 2025
Configuring OpenRouter language models Integrations ai	0	1037	10 de Dezembro de 2024
How to implement Mistral with Embeddings Support related-topics , ai	6	228	11 de Maio de 2025
Inquiry About AI Plugin Options Support ai	7	133	24 de Novembro de 2025

Como usar modelos de IA com tokenizadores customizados

Tópicos relacionados