Together AI
google/gemma-4-31b-it model retirement
Stopped workingSubscribe to Together AIShare
Together AI retired model google/gemma-4-31b-it on September 14, 2026. Calls specifying this model fail since that date. The change affects any requests sent to the Together AI API targeting this identifier.
What to use instead of google/gemma-4-31b-it
Move to zai-org/GLM-5.3-Flash, which Together AI recommends as the replacement.
In Together AI's words
Removal date: 2026-09-14; Model: google/gemma-4-31B-it; Recommended replacement: zai-org/GLM-5.3-Flash; Supported by on-demand dedicated endpoints: Yes
Facts
- API
- Together AI
api.together.xyz - Model
google/gemma-4-31b-it- Released
- April 2, 2026 (models.dev)
- Kind
- Model retirement
- Takes effect
- September 14, 2026
- Calls after that day
- Fail
- Source
- Deprecations page
- Last updated
- Verified
- , against Together AI's own pages
Questions
When does google/gemma-4-31b-it shut down?
Calls to the model google/gemma-4-31b-it on Together AI stopped working on September 14, 2026.
What happens to calls after September 14, 2026?
They have failed since September 14, 2026.
When was google/gemma-4-31b-it released?
google/gemma-4-31b-it was released on April 2, 2026, 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.
google/gemma-4-31b-it on other APIs 2
-
Together AI
pearl-ai/gemma-4-31b-itModel retirement Stopped working -
Novita AI
google/gemma-4-31b-itModel retirement Stops workingin 20 days
More from Together AI 10
-
deepseek-ai/deepseek-v4-proModel retirement Stopped working -
moonshotai/kimi-k2.7-codeModel retirement Stopped working -
nvidia/nemotron-3-ultra-550b-a55bModel retirement Stopped working -
pearl-ai/gemma-4-31b-itModel retirement Stopped working -
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning-bf16Model retirement Stopped working -
nvidia/nvidia-nemotron-3-super-120b-a12b-bf16Model retirement Stopped working -
qwen/qwen3.5-35b-a3b-baseModel retirement Stopped working -
intfloat/multilingual-e5-large-instructModel retirement Stopped working -
openai/gpt-oss-20bModel retirement Stopped working -
thinkingmachines/inkling-smallModel 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.