By tencent ยท Updated 2026-11-22 ยท Open weights ยท custom licenseReads imageslong-context / RAGreads images / documentscheap high-volumestep-by-step reasoningopen weights / self-hostable
Tencent: Hy4 flash is best suited for long-context / RAG, reads images / documents, cheap high-volume, step-by-step reasoning, open weights / self-hostable. Access it through one Appmo API key alongside 467 other models โ Appmo's optimizer can route to it automatically when it is the best value for a request. Weights: huggingface.co/tencent/Hy4-flash.
Not enough posts yet to judge community sentiment. 5 Hacker News stories (187 points) ยท 19 new GitHub repos mentioning it. How we measure this.
curl https://appmo.com/v1/chat/completions \
-H "Authorization: Bearer $APPMO_KEY" \
-d '{"model":"tencent/hy4-flash","messages":[{"role":"user","content":"Hello"}]}'How much does Tencent: Hy4 flash cost?
Tencent: Hy4 flash costs $0.035 per 1M input tokens and $0.1405 per 1M output tokens ($0.061 blended), as of 2026-11-22. Prices are expected to fall as competition increases.
What is Tencent: Hy4 flash best for?
Tencent: Hy4 flash is best suited for long-context / RAG, reads images / documents, cheap high-volume, step-by-step reasoning, open weights / self-hostable.
How do I use Tencent: Hy4 flash?
Send requests to Appmo's OpenAI-compatible endpoint with model id "tencent/hy4-flash" and one API key โ no separate provider account needed.
Is Tencent: Hy4 flash open source?
Yes โ its weights are public on Hugging Face (tencent/Hy4-flash) under a custom license; check the license terms before commercial use.
Can Tencent: Hy4 flash create images?
No. Tencent: Hy4 flash can read and understand images, but it only writes text.