[torrresagus]

What changed in the API

Notice groq-retired-models-2026-08 about api.groq.com's API describes this change, published 2026-06-17 and effective 2026-08-16. It is not signed by api.groq.com: it was read from a file the person running Tiltmeter trusts.

The notice's summary:

Groq shut down llama-3.3-70b-versatile and llama-3.1-8b-instant on August 16, 2026 for free and
developer-tier keys, after the Llama, Mixtral, Gemma, DeepSeek and Qwen models it retired between
2024 and 2026. Requests that name them return an error. Each change names Groq's recommended
replacement or, where that replacement was retired too, its current successor. Source:
https://console.groq.com/docs/deprecations

POST https://api.groq.com/openai/v1/chat/completions

Severity	Change	Path	Before	After
breaking	value_retired	$.model	llama-3.3-70b-versatile	(no longer accepted), send openai/gpt-oss-120b instead
breaking	value_retired	$.model	llama-3.1-8b-instant	(no longer accepted), send openai/gpt-oss-20b instead
breaking	value_retired	$.model	meta-llama/llama-4-scout-17b-16e-instruct	(no longer accepted), send openai/gpt-oss-120b instead
breaking	value_retired	$.model	meta-llama/llama-4-maverick-17b-128e-instruct	(no longer accepted), send openai/gpt-oss-120b instead

The notice's notes:

llama-3.3-70b-versatile: Shut down on August 16, 2026: requests that name it return an error. The
shutdown applies to free and developer-tier keys; Groq enterprise customers with a committed-spend
contract keep the model. Groq recommends openai/gpt-oss-120b (its other suggestion,
qwen/qwen3.6-27b, was itself shut down on September 14, 2026). openai/gpt-oss-120b and openai/gpt-
oss-20b are reasoning models: they reason before they answer, and that reasoning counts toward the
completion token limit.
llama-3.1-8b-instant: Shut down on August 16, 2026: requests that name it return an error. The
shutdown applies to free and developer-tier keys; Groq enterprise customers with a committed-spend
contract keep the model. openai/gpt-oss-20b is Groq's recommended replacement. openai/gpt-oss-120b
and openai/gpt-oss-20b are reasoning models: they reason before they answer, and that reasoning
counts toward the completion token limit.
meta-llama/llama-4-scout-17b-16e-instruct: Shut down on July 17, 2026. The shutdown applies to free
and developer-tier keys; Groq enterprise customers with a committed-spend contract keep the model.
Groq recommends openai/gpt-oss-120b (its other suggestion, qwen/qwen3.6-27b, was itself shut down on
September 14, 2026). Neither replacement reads images.
meta-llama/llama-4-maverick-17b-128e-instruct: Shut down on March 9, 2026. openai/gpt-oss-120b is
Groq's recommended replacement; it does not read images.

What this pull request changes

Changed the default Groq chat model in GroqModels.DEFAULT_MODEL from the retired llama-3.1-8b-instant to Groq's recommended replacement, openai/gpt-oss-20b. Users who picked a model themselves keep that choice. The fallback model list was left as it is.

• app/src/main/java/helium314/keyboard/latin/utils/GroqModels.kt

Left for a human

• GroqModels.AVAILABLE_MODELS (used as the picker list and as the fallback when fetching /models fails) still lists the four retired models: llama-3.3-70b-versatile, llama-3.1-8b-instant, meta-llama/llama-4-scout-17b-16e-instruct and meta-llama/llama-4-maverick-17b-128e-instruct. Their replacements are already in the list, and enterprise keys can still use these models, so whether to remove them is the maintainers' call.
• Users whose saved KEY_GROQ_MODEL preference names a retired model will now get API errors. The saved preference was not migrated because it is user configuration.
• openai/gpt-oss-20b is a reasoning model, and its reasoning counts toward the completion token limit. Groq requests always send max_tokens = getCloudMaxTokens(), and isReasoningOrNewerModel() does not recognize gpt-oss, so the request still includes temperature 0.1. Check that the configured token limit leaves room for the answer, or answers may come back cut off or empty.

---

This pull request was generated automatically by Tiltmeter from the unsigned notice groq-retired-models-2026-08 about api.groq.com's API. An AI agent (Claude Code with claude-opus-5-5) wrote these edits; review them like any other change.