SambaNova Cloud
llama-4-maverick-17b-128e-instruct model retirement
Stopped workingSubscribe to SambaNova CloudShare
On June 9, 2026, SambaNova Cloud retired model llama-4-maverick-17b-128e-instruct. Calls targeting this model fail since that day. Requests are not redirected to a newer version.
What to use instead of llama-4-maverick-17b-128e-instruct
Move to gemma-4-31B-it, which is recommended by SambaNova.
In SambaNova Cloud's words
Model ID: Llama-4-Maverick-17B-128E-Instruct; Removal date: 6/9/2026; Recommended replacement: gemma-4-31B-it
Facts
- API
- SambaNova Cloud
api.sambanova.ai - Model
llama-4-maverick-17b-128e-instruct- Kind
- Model retirement
- Takes effect
- June 9, 2026
- Calls after that day
- Fail
- Source
- Deprecations page
- Last updated
- Verified
- , against SambaNova Cloud's own pages
Questions
When does llama-4-maverick-17b-128e-instruct shut down?
Calls to the model llama-4-maverick-17b-128e-instruct on SambaNova Cloud stopped working on June 9, 2026.
What happens to calls after June 9, 2026?
They have failed since June 9, 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.
llama-4-maverick-17b-128e-instruct on other APIs 1
-
Groq
meta-llama/llama-4-maverick-17b-128e-instructModel retirement Stopped working
More from SambaNova Cloud 10
-
deepseek-r1-distill-llama-70bModel retirement Stopped working -
deepseek-v3.1-terminusModel retirement Stopped working -
e5-mistral-7b-instructModel retirement Stopped working -
qwen3-235b-a22b-instruct-2507Model retirement Stopped working -
qwen3-32bModel retirement Stopped working -
deepseek-r1-0528Model retirement Stopped working -
deepseek-v3-0324Model retirement Stopped working -
meta-llama-3.1-8b-instructModel retirement Stopped working -
minimax-m2.5Model retirement Stopped working -
gemma-3-12b-itModel 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.