Okapi Okapi

DeepInfra qwen/qwen3.8-2.4t-a95b model retirement

Stopped workingSubscribe to DeepInfraShare

DeepInfra retired the model qwen/qwen3.8-2.4t-a95b on October 12, 2026. Calls specifying this model fail since that date. Any requests sent to api.deepinfra.com targeting this model are no longer processed.


What to use instead of qwen/qwen3.8-2.4t-a95b

Move to zai-org/GLM-5.3, which is recommended by DeepInfra.


Facts

API
DeepInfra api.deepinfra.com
Model
qwen/qwen3.8-2.4t-a95b
Released
August 12, 2026 (models.dev)
Kind
Model retirement
Takes effect
October 12, 2026
Calls after that day
Fail
Source
Model list
Last updated
Verified
, against DeepInfra's own pages

Questions

When does qwen/qwen3.8-2.4t-a95b shut down?

Calls to the model qwen/qwen3.8-2.4t-a95b on DeepInfra stopped working on October 12, 2026.

What happens to calls after October 12, 2026?

They have failed since October 12, 2026.

When was qwen/qwen3.8-2.4t-a95b released?

qwen/qwen3.8-2.4t-a95b was released on August 12, 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.


qwen/qwen3.8-2.4t-a95b on other APIs 2

  1. Together AI qwen/qwen3.8-2.4t-a95bModel retirement Stopped working
  2. No date DeepInfra qwen/qwen3.8-2.4t-a95bModel retirement Deprecated, no date

More from DeepInfra 10

  1. black-forest-labs/flux.1-kontext-devModel retirement Stopped working
  2. deepseek-ai/janus-pro-1bModel retirement Stopped working
  3. deepseek-ai/janus-pro-7bModel retirement Stopped working
  4. nvidia/nemotron-3-nano-30b-a3bModel retirement Stopped working
  5. nvidia/nvidia-nemotron-nano-9b-v2Model retirement Stopped working
  6. qwen/qwen-image-editModel retirement Stopped working
  7. tencent/hy3Model retirement Stopped working
  8. inclusionai/ling-3.0-flash-finModel retirement Stopped working
  9. qwen/qwen3-maxModel retirement Stopped working
  10. qwen/qwen3-max-thinkingModel 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.

Add to GitHub

Get notifications

Subscribe to DeepInfra

Hear about each breaking change DeepInfra announces, in your feed reader, calendar or Slack.

1What to follow

Provider
  • DeepInfra

2Where to get it

/breaking-changes/feed.xml

Paste it into Feedly, Inoreader, NetNewsWire or any reader. New changes show up as they're announced.