Releases·Model Release
GitHub rolls out Tiers of Auto with Efficiency, Balance, and Intelligence modes for Copilot model selection
On September 14, 2026, GitHub announced Tiers of Auto, letting developers pick cost/quality tradeoffs in Copilot's Auto model selection. Three modes — Efficiency, Balance, Intelligence — are available across VS Code, CLI, and the Copilot App. This marks GitHub's first step toward personalized, configurable intelligent model routing.
68 engagement1 sourceMon, Sep 14, 2026, 12:00 AM
AI Overview
- GitHub launched Tiers of Auto on September 14, 2026, letting developers pick cost-quality tradeoffs in Auto model selection across VS Code, CLI, and Copilot App.
- Tiers offer three modes: Efficiency for low-cost fast tasks, Balance weighing cost, quality, and latency, and Intelligence prioritizing quality for complex tasks.
- Auto added new models to its mixes including GPT 6 Astra, GPT 5.6 series, and Opus 5, with more models coming.
- Adaptive multi-turn support lets Auto reevaluate complexity throughout a session, expanding beyond prior evaluation points at conversation start and after compaction.
- GitHub improved the Copilot App UX to explain why Auto chose a model, with further enhancements planned.
Related
Social·Community
Users compare Claude Sonnet 5, Gemini 3.7 Flash, Qwen3.8-Max for daily workflows
A Reddit user compares Claude Sonnet 5, Gemini 3.7 Flash, and Qwen3.8-Max as daily drivers, noting Claude's strength in deep coding and agentic workflows, and Gemini's speed-to-cost ratio. The post reflects community impressions of recently released models.
A Reddit user compares Claude Sonnet 5, Gemini 3.7 Flash, and Qwen3.8-Max as daily drivers, noting Claude's strength in deep coding and agentic workflows, and Gemini's speed-to-cost ratio. The post reflects community impressions of recently released models.
6 engagement1 sourceredditFri, Aug 21, 2026, 10:32 PM
Releases·Tool Release
langchain-openai==1.6.1 ships async tools, Azure AD fix, gpt-5.6-sol routing
LangChain shipped langchain-openai 1.6.1 on GitHub, fixing routing of gpt-5.6-sol to the responses API and adding support for async tools, configuration_update, and Azure AD auth with OpenAI 3.8. The release matters to developers building on OpenAI models through LangChain, since it also corrects reasoning_effort_levels metadata for gpt-5 and gpt-5.1.
LangChain shipped langchain-openai 1.6.1 on GitHub, fixing routing of gpt-5.6-sol to the responses API and adding support for async tools, configuration_update, and Azure AD auth with OpenAI 3.8. The release matters to developers building on OpenAI models through LangChain, since it also corrects reasoning_effort_levels metadata for gpt-5 and gpt-5.1.
0 engagement1 sourcegithubTue, Sep 8, 2026, 02:20 PM
Releases·Tool Release
Ollama v0.34.0 runs local models in ChatGPT Desktop, improves Apple Silicon structured output
Ollama shipped v0.34.0, letting users run Ollama models directly in ChatGPT Desktop, with setup available from the Ollama app on MacOS. The release also improves structured output performance on Apple Silicon and adds OpenAI-compatible client tool search and response compaction support.
Ollama shipped v0.34.0, letting users run Ollama models directly in ChatGPT Desktop, with setup available from the Ollama app on MacOS. The release also improves structured output performance on Apple Silicon and adds OpenAI-compatible client tool search and response compaction support.
17 engagement1 sourcegithubSat, Sep 5, 2026, 11:49 PM
Social·Community
Users compare GitHub Copilot with GPT 5.6 Sol vs. Claude with Opus 5
A developer shares an early hands-on comparison of GitHub Copilot using GPT 5.6 Sol via a custom endpoint versus Claude with Opus 5. The post focuses on harness differences rather than raw model quality, noting both setups were slow on a simple task. The user had switched to Claude months prior and found it great.
A developer shares an early hands-on comparison of GitHub Copilot using GPT 5.6 Sol via a custom endpoint versus Claude with Opus 5. The post focuses on harness differences rather than raw model quality, noting both setups were slow on a simple task. The user had switched to Claude months prior and found it great.
32 engagement2 sourcesredditWed, Aug 12, 2026, 02:19 AM
Releases·Model Release
Kimi K3 open-weight model launches in GitHub Copilot
Kimi K3, an open-weight model, is now generally available in GitHub Copilot, offering frontier-level agentic coding abilities at a highly cost-effective price. The release expands Copilot's model lineup with a competitive open-weight option for developers.
Kimi K3, an open-weight model, is now generally available in GitHub Copilot, offering frontier-level agentic coding abilities at a highly cost-effective price. The release expands Copilot's model lineup with a competitive open-weight option for developers.
313 engagement1 sourceredditThu, Aug 6, 2026, 05:30 PM