Together AI
mistralai/mistral-7b-instruct-v0.3 model retirement
RedirectedSubscribe to Together AIShare
On November 6, 2026, Together AI retired model mistralai/mistral-7b-instruct-v0.3. Calls targeting this model are now automatically redirected to a newer version rather than failing. Together AI answers requests from mistralai/Ministral-3-14B-Instruct-2512.
What to use instead of mistralai/mistral-7b-instruct-v0.3
Move to mistralai/Ministral-3-14B-Instruct-2512, which is recommended by Together AI.
In Together AI's words
Original model: mistralai/Mistral-7B-Instruct-v0.3; Redirects to: mistralai/Ministral-3-14B-Instruct-2512; Notes: Same lineage, upgraded version
Facts
- API
- Together AI
api.together.xyz - Model
mistralai/mistral-7b-instruct-v0.3- Released
- May 22, 2024 (models.dev)
- Kind
- Model retirement
- Takes effect
- November 6, 2026
- Calls after that day
- Answered by a newer version
- Source
- Deprecations page
- Last updated
- Verified
- , against Together AI's own pages
Questions
When does this change take effect?
Since November 6, 2026, Together AI answers calls to the model mistralai/mistral-7b-instruct-v0.3 from a newer version.
What happens to calls after November 6, 2026?
Together AI answers them from a newer version. The response can differ from what your code expects.
When was mistralai/mistral-7b-instruct-v0.3 released?
mistralai/mistral-7b-instruct-v0.3 was released on May 22, 2024, 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.
mistralai/mistral-7b-instruct-v0.3 on other APIs 2
-
DeepInfra
mistralai/mistral-7b-instruct-v0.3Model retirement Stopped working -
No date
NVIDIA
mistralai/mistral-7b-instruct-v0.3Model retirement Deprecated, no date
More from Together AI 10
-
google/gemma-4-31b-itModel retirement Stopped working -
deepseek-ai/deepseek-v4-flash-0731Model retirement Stopped working -
meta-llama/llama-3.3-70b-instruct-turboModel retirement Stopped working -
qwen/qwen3.8-2.4t-a95bModel retirement Stopped working -
deepseek-r1Model retirement Redirected -
deepseek-v3Model retirement Redirected -
deepseek-v3-0324Model retirement Redirected -
kimi-k2Model retirement Redirected -
minimaxai/minimax-m3Model retirement Stopped working -
thinkingmachines/inklingModel 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.