Novita AI
qwen/qwen3-8b-fp8 model retirement
Stopped workingSubscribe to Novita AIShare
Calls to the model qwen/qwen3-8b-fp8 on Novita AI stopped working on January 8, 2026.
What to use instead of qwen/qwen3-8b-fp8
Move to qwen/qwen3.8-max. qwen/qwen3-next-80b-a3b-instruct, named first, stopped working on 2026-10-09.
In Novita AI's words
Deprecated model: qwen/qwen3-8b-fp8; Recommended alternatives: qwen/qwen3-next-80b-a3b-instruct
Facts
- API
- Novita AI
api.novita.ai - Model
qwen/qwen3-8b-fp8- Kind
- Model retirement
- Takes effect
- January 8, 2026
- Calls after that day
- Fail
- Source
- Model notices
- Last updated
- Verified
- , against Novita AI's own pages
Questions
When does qwen/qwen3-8b-fp8 shut down?
Calls to the model qwen/qwen3-8b-fp8 on Novita AI stopped working on January 8, 2026.
What happens to calls after January 8, 2026?
They have failed since January 8, 2026.
How do I find the calls this hits in my code?
Okapi reads your repository, finds the calls this change hits and opens a pull request with the fix. It never commits to your branches: every change comes as a pull request you review.
More from Novita AI 10
-
baichuan/baichuan-m2-32bModel retirement Stopped working -
deepseek/deepseek-ocrModel retirement Stopped working -
deepseek/deepseek-r1-0528-qwen3-8bModel retirement Stopped working -
deepseek/deepseek-r1-distill-qwen-14bModel retirement Stopped working -
deepseek/deepseek-r1-distill-qwen-32bModel retirement Stopped working -
google/gemma-3-12b-itModel retirement Stopped working -
gryphe/mythomax-l2-13bModel retirement Stopped working -
meta-llama/llama-3.2-3b-instructModel retirement Stopped working -
paddlepaddle/paddleocr-vlModel retirement Stopped working -
qwen/qwen3-reranker-8bModel retirement Stopped working
You fixed this one by hand.
Okapi opens the pull request with the fix before the next breaking change reaches your code.