2026-07-03 12:03 UTC · groq/deprecations · live page
Summary by fallback. It may be wrong; the diff below is what actually changed. Raw: 793919aed2-groq-deprecations.diff · commit 793919aed2
diff --git a/snapshots/groq/deprecations.md b/snapshots/groq/deprecations.mdindex f62db64..e1f60e4 100644--- a/snapshots/groq/deprecations.md+++ b/snapshots/groq/deprecations.md@@ -39,4 +39,18 @@ When a model is marked for deprecation, we follow this standardized process: - Consider designing your systems to be model-agnostic where possible +In line with our commitment to bringing you cutting-edge models, on June 17, 2026, we emailed users to announce the deprecation of `llama-3.1-8b-instant` and `llama-3.3-70b-versatile`. We recommend migrating to `openai/gpt-oss-20b` (for Llama 3.1 8B Instant) and `openai/gpt-oss-120b` or `qwen/qwen3.6-27b` (for Llama 3.3 70B Versatile), which deliver exceptional performance with faster inference. This deprecation applies to free and developer-tier usage; enterprise customers with a committed-spend contract are not affected.++| Deprecated Model | Shutdown Date | Recommended Replacement Model ID |+|---|---|---|+| `llama-3.1-8b-instant` | 08/16/26 | `openai/gpt-oss-20b` |+| `llama-3.3-70b-versatile` | 08/16/26 | `openai/gpt-oss-120b` or `qwen/qwen3.6-27b` |++In line with our commitment to bringing you cutting-edge models, on June 17, 2026, we emailed users to announce the deprecation of `qwen/qwen3-32b` and `meta-llama/llama-4-scout-17b-16e-instruct`. We recommend migrating to `openai/gpt-oss-120b` (for Qwen 3 32B) and `openai/gpt-oss-120b` or `qwen/qwen3.6-27b` (for Llama 4 Scout 17B), which deliver exceptional performance with faster inference. This deprecation applies to free and developer-tier usage; enterprise customers with a committed-spend contract are not affected.++| Deprecated Model | Shutdown Date | Recommended Replacement Model ID |+|---|---|---|+| `qwen/qwen3-32b` | 07/17/26 | `openai/gpt-oss-120b` |+| `meta-llama/llama-4-scout-17b-16e-instruct` | 07/17/26 | `openai/gpt-oss-120b` or `qwen/qwen3.6-27b` |+ In line with our commitment to bringing you cutting-edge models, on March 23, 2026, we emailed users to announce the deprecation of `moonshotai/kimi-k2-instruct-0905` in favor of `openai/gpt-oss-120b`. The GPT-OSS 120B model delivers exceptional reasoning performance with faster inference, enabling your applications to harness state-of-the-art AI capabilities with unparalleled speed on our platform.