Both the Python and TypeScript SDKs add guidance for Qwen 3.8 reasoning in chat completions. The Python SDK also updates dependency lock files for known vulnerabilities.
Changelog
npx @buildinternet/releases get groq-changelogqwen/qwen3.6-27b is deprecated with a shutdown date of September 14, 2026, affecting free and developer-tier usage. Migrate to qwen/qwen3.8-27b.
A batch that names a model the organization cannot use now fails at validation before any request runs, with status "failed" and request_counts.total: 0. Errors are listed per denied model in errors.data[].
groq/compound and groq/compound-mini were decommissioned on September 21, 2026, and requests to these model IDs now return errors.
Organizations can now be subject to separate per-minute limits on input tokens and output tokens. The Limits page shows an "X in / Y out" breakdown on hover when configured.
The built-in browser_automation tool for Compound systems has been retired. Anchor Browser integration is the replacement.
qwen/qwen3-32b and meta-llama/llama-4-scout-17b-16e-instruct were shut down on July 17, 2026, with openai/gpt-oss-120b recommended as the replacement.
llama-3.1-8b-instant and llama-3.3-70b-versatile are deprecated and shut down on August 16, 2026. Migrate to openai/gpt-oss-20b and openai/gpt-oss-120b respectively.
qwen/qwen3.6-27b is now available on GroqCloud, a 27B multimodal model with agentic coding and reasoning, switchable thinking and non-thinking modes, and a 131K-token context window.
Both SDKs now support header configuration through environment variables, and the Python SDK adds pydantic iterator validation. The release also includes API spec updates and Node 24/26 compatibility and CI changes.
MiniMax M2.7 is now available on GroqCloud for Enterprise customers, a 229B-parameter MoE model with interleaved thinking built for agentic workflows and coding with a 196K-token context window.
Python SDK v1.2.0 preserves hardcoded query params, improves multipart file-copy performance, and adds binary request streaming with a custom JSON encoder; Python 3.9 is deprecated. TypeScript SDK v1.1.2 restores streaming in defaultParseResponse and fixes an abort-signal memory leak and minimatch CVEs.
Two new voices are available for the Orpheus Arabic Saudi TTS model, with Abdullah now the default. The model now supports six voices total.
MiniMax M2.5 general-purpose model and Qwen3-VL 32B Instruct vision-language model are now available on GroqCloud for Enterprise customers. Python SDK v1.2.0 preserves hardcoded query params when merging with user params, ensures file data is sent as a single parameter, and improves multipart request file-copy performance. TypeScript SDK v1.1.2 restored streaming support in defaultParseResponse, fixed an abort-signal memory leak, and patched a minimatch CVE.
Text-to-speech has migrated platform-wide from PlayAI to Orpheus models from Canopy Labs, which offer enhanced expressiveness with vocal direction controls, faster inference, and improved audio quality. Users still on playai-tts or playai-tts-arabic must migrate before the shutdown date listed on the deprecations page.
Groq now offers pre-built MCP Connectors for Google Workspace applications—Gmail, Google Calendar, and Google Drive—with zero configuration and OAuth 2.0 authentication. Available tools include email search and reading, calendar event viewing, and file search and access.
OpenAI's GPT-OSS-Safeguard 20B, an open-weight reasoning model for safety classification, is now available on Groq. The model offers a 131K context window, up to 65K output tokens, ~1000 TPS throughput, prompt caching for 50% cost savings, and supports tool use, code execution, and multiple response formats for Trust & Safety content moderation and policy-based classification tasks.
Automatic prompt caching is now available for openai/gpt-oss-120b, reducing cached input token costs to $0.075/M (from $0.15/M), lowering latency, and increasing effective rate limits since cached tokens do not count toward limits. Python SDK updated to v0.33.0 and TypeScript SDK to v0.34.0 with improved prompt caching support and annotation/citation support for chat completion messages and streamed deltas.
Both SDKs add annotation and citation support to chat completion messages and streamed deltas, with improved prompt caching.
Automatic prompt caching is now available for openai/gpt-oss-20b, reducing cached input token costs by 50% (from $0.075/M to $0.037/M) while lowering latency through automatic prefix matching. No configuration required.


