Together AI
moonshotai/kimi-k2-instruct model retirement
Stopped workingSubscribe to Together AIShare
Together AI retired model moonshotai/kimi-k2-instruct on July 29, 2026. Calls targeting this model fail since that date. The removal took effect without an earlier announced date in the records.
What to use instead of moonshotai/kimi-k2-instruct
The initial suggested replacement moonshotai/Kimi-K2.6 also stopped working on August 19, 2026. No replacement that holds is named yet.
In Together AI's words
Removal date: 2026-07-29; Model: moonshotai/Kimi-K2-Instruct; Suggested replacement: moonshotai/Kimi-K2.6
Facts
- API
- Together AI
api.together.xyz - Model
moonshotai/kimi-k2-instruct- Released
- July 11, 2025 (models.dev)
- Kind
- Model retirement
- Takes effect
- July 29, 2026
- Calls after that day
- Fail
- Source
- Deprecations page
- Last updated
- Verified
- , against Together AI's own pages
Questions
When does moonshotai/kimi-k2-instruct shut down?
Calls to the model moonshotai/kimi-k2-instruct on Together AI stopped working on July 29, 2026.
What happens to calls after July 29, 2026?
They have failed since July 29, 2026.
When was moonshotai/kimi-k2-instruct released?
moonshotai/kimi-k2-instruct was released on July 11, 2025, according to models.dev.
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.
moonshotai/kimi-k2-instruct on other APIs 2
-
Groq
moonshotai/kimi-k2-instructModel retirement Stopped working -
DeepInfra
moonshotai/kimi-k2-instructModel retirement Stopped working
More from Together AI 10
-
deepseek-ai/deepseek-r1-0528Model retirement Stopped working -
deepseek-ai/deepseek-r1-distill-llama-70b-131kModel retirement Stopped working -
deepseek-ai/deepseek-r1-distill-llama-70b-32kModel retirement Stopped working -
deepseek-ai/deepseek-v3Model retirement Stopped working -
deepseek-ai/deepseek-v3-0324Model retirement Stopped working -
deepseek-ai/deepseek-v3-baseModel retirement Stopped working -
deepseek-ai/deepseek-v3.1-baseModel retirement Stopped working -
google/gemma-3-12b-itModel retirement Stopped working -
google/gemma-3-12b-it-vlmModel retirement Stopped working -
google/gemma-3-12b-ptModel 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.