Deprecation refers to the process of retiring older models or endpoints in favor of hosting better models with better capabilities for you to leverage. When we announce that a model or endpoint is being deprecated, we will provide a shutdown date on which the model or endpoint will no longer be accessible. As such, your applications relying on Groq may need occasional updates to continue working.
Once a model is announced as deprecated, make sure to migrate usage to a recommended replacement before the shutdown date to avoid failing requests. All API deprecations along with recommended replacements are listed below.
We ship fast so you can build fast with access to the latest and greatest models, while also providing a distinction between preview models and production models. Generally, models we host start off in preview and either graduate to production based on demand or get replaced by a production model with similar or better capabilities.
When a model is marked for deprecation, we follow this standardized process:
Announcement Phase:
Transition Phase:
Automatic Upgrade Phase (when applicable):
End-of-Life:
In line with our commitment to bringing you cutting-edge models, on April 7, 2025, we emailed users to announce the deprecation of several older preview models in favor of Meta's Llama 4 suite. The new Llama 4 Scout and Maverick models deliver exceptional multimodal performance that outpaces our previous offerings, enabling your applications to harness state-of-the-art AI capabilities with unparalleled speed on our platform.
| Deprecated Model | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
llama-3.2-1b-preview | 04/14/25 | llama-3.1-8b-instant |
llama-3.2-3b-preview | 04/14/25 | llama-3.1-8b-instant |
llama-3.2-11b-vision-preview | 04/14/25 | meta-llama/llama-4-scout-17b-16e-instruct |
llama-3.2-90b-vision-preview | 04/14/25 | meta-llama/llama-4-scout-17b-16e-instruct |
deepseek-r1-distill-qwen-32b | 04/14/25 | qwen-qwq-32b |
qwen-2.5-32b | 04/14/25 | qwen-qwq-32bmeta-llama/llama-4-scout-17b-16e-instruct |
qwen-2.5-coder-32b | 04/14/25 | qwen-qwq-32bmeta-llama/llama-4-maverick-17b-128e-instruct |
llama-3.3-70b-specdec | 04/14/25 | meta-llama/llama-4-scout-17b-16e-instructllama-3.3-70b-versatile |
deepseek-r1-distill-llama-70b-specdec | 04/14/25 | deepseek-r1-distill-llama-70bdeepseek-r1-distill-qwen-32b |
On March 17, 2025, we emailed all users of the deepseek-r1-distill-llama-70b-specdec model that we would be deprecating this model ID in favor of our standard DeepSeek R1 Distill Llama 70B model and the DeepSeek
R1 Distill Qwen 32B reasoning model, both of which are more popular with our users for their performance.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
deepseek-r1-distill-llama-70b-specdec | 03/24/25 | deepseek-r1-distill-llama-70bdeepseek-r1-distill-qwen-32b |
On March 5, 2025, we emailed all users of the mixtral-8x7b-32768 model that we would be deprecating this model ID in favor of newer, more performant models. The recommended replacement models offer superior
multilingual capabilities and performance for various tasks from text generation to translation.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
mixtral-8x7b-32768 | 03/20/25 | mistral-saba-24bllama-3.3-70b-versatile |
On December 6, 2024, in partnership with Meta, we released llama-3.3-70b-versatile and llama-3.3-70b-specdec, and notified users that we would deprecate their 3.1 counterparts in favor of hosting Llama 3.3
with significant quality improvements for a better experience.
To facilitate a smooth transition, we will maintain the current llama-3.1-70b-versatile and llama-3.1-70b-specdec model IDs until December 20, 2024. At that time, requests to these model IDs will automatically
upgrade to their respective 3.3 versions. Beginning January 24, 2025, requests to both 3.1 model IDs will return errors.
While these new models deliver improved quality, they may produce different responses than their predecessors. We recommend migrating to explicitly using llama-3.3-70b-versatile and llama-3.3-70b-specdec before
December 20, 2024, for testing.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
llama-3.1-70b-versatile | 01/24/25 | llama-3.3-70b-versatile |
llama-3.1-70b-specdec | 01/24/25 | llama-3.3-70b-specdec |
On January 6th, we deprecated our preview versions of Llama 3 fine-tuned for tool use, llama3-groq-8b-8192-tool-use-preview and llama3-groq-70b-8192-tool-use-preview, from GroqCloud™ in favor of
transitioning users to our production-ready llama-3.30-70b-versatile model.
Users of the tool use models were notified about the upcoming deprecation via email. The recommended replacement model, llama-3.3-70b-versatile, offers superior tool use capabilities and we strongly encourage
users to migrate applications to this model for improved reliability and performance.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
llama3-groq-8b-8192-tool-use-preview | 1/6/25 | llama-3.3-70b-versatile |
llama3-groq-70b-8192-tool-use-preview | 1/6/25 | llama-3.3-70b-versatile |
On December 11, 2024, we emailed all Gemma 7B users that we would deprecate it in favor of keeping the Gemma 9B model as it offers better performance.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
gemma-7b-it | 12/18/24 | gemma2-9b-it |
In November 2024, we emailed all Llama 3.2 90B Text Preview users that we would deprecate it in favor of hosting the Llama 3.2 90B Vision Preview model for vision capabilities.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
llama-3.2-90b-text-preview | 11/25/24 | llama-3.2-90b-vision-previewllama-3.1-70b-versatile |
In September 2024, we made Meta's Llama 3.2 vision models available on GroqCloud and emailed all LLaVA 1.5 7B and Llama 3.2 11B Text Preview users that we would deprecate it in favor of hosting Llama 3.2 11B Vision for better performance and more robust vision capabilities.
| Model ID | Shutdown Date | Recommended Replacement Model ID |
|---|---|---|
llava-v1.5-7b-4096-preview | 10/28/24 | llama-3.2-11b-vision-preview |
llama-3.2-11b-text-preview | 10/28/24 | llama-3.2-11b-vision-previewllama-3.1-8b-instant |