.Anthropic’s Claude Opus 4.6 is now generally available and rolling out in GitHub Copilot. Early testing shows Claude Opus 4.6 👇 ➡️ Excels in agentic coding ➡️ Performs well with hard tasks requiring planning and tool calling Try it out yourself in Visual Studio Code.

@AnthropicAI @code Why is the context window still capped at 128k?

@AnthropicAI @code Wow super fast, now can you change Opus 4.5 to 2x or 1.5x to compare...

@AnthropicAI @code Does it get the full 1M context window? 🥺

@AnthropicAI @code How about ChatGPT-Codex-5.3????

@AnthropicAI @code Opus 4.6 + Codex in the same Copilot? This is the real competition — not Anthropic vs OpenAI, but models competing to be YOUR default in the same IDE. Best day to be a developer. Worst day to pick just one model.

@AnthropicAI @code When Codex 5.3?

@AnthropicAI @code been running opus in claude code all day. the planning on multi-file refactors is noticeably better than 4.5. havent tried it in copilot yet tho, wonder if the agentic mode feels different there vs terminal

@AnthropicAI @code What the context window on GitHub Copilot ? 100k ? 200k ? Or did you dare give us the full model?

@AnthropicAI @code codex 5.3?

@AnthropicAI @code Good speed, well coordinated.

@AnthropicAI @code opus 4.6 in copilot is basically giving your ide a senior dev title. if the tool calling is actually reliable i’m essentially just a glorified code reviewer now

@AnthropicAI @code Will it have 1X discount like last time with Opus 4.5? 👀

@AnthropicAI @code We also need gpt-5.3-codex plz

@AnthropicAI @code Opus 4.6 catching its own mistakes in codebases is huge... but what about real-world mistakes? Agents ship flawless logic yet still fail when the physical button doesn't light up or the shelf is empty. That's the perception gap. Give them eyes first 🦞

@AnthropicAI @code Does it offers the 1M context window?

@AnthropicAI @code Cmon guys, the context size should not be limited to 128k...

@AnthropicAI @code 5.3 codex when?

@AnthropicAI @code How can I test GPT-5.3 Codex in the VS Code IDE? Has it been enabled there as well?

@AnthropicAI @code Waiting for GPT-5.3-Codex. 🙏

@AnthropicAI @code Exciting update! Can't wait to see what Claude Opus 4.6 can do. Time to put those agentic coding skills to the test in GitHub Copilot

@AnthropicAI @code With the copilot extension you have the perfect abstraction layer for AI agents, def will give it a try tomorrow

@AnthropicAI @code so fast

@AnthropicAI @code Kal use karunga

@AnthropicAI @code Slick 😎

@AnthropicAI @code the integration layer is becoming more important than the model itself. having opus, codex, and gemini all available in one copilot means the workflow determines which model you reach for, not vendor lock in. curious how context sharing works when switching between them mid task.

@AnthropicAI @code So devs now have zero excuses left or we’ll just start blaming Claude for every bug

@AnthropicAI @code This is a strong move for developer productivity 💻 Agentic coding + deep planning inside Copilot means fewer handoffs and faster shipping. Excited to see how Opus 4.6 raises the bar for complex builds

@AnthropicAI @code It's about time Copilot brings other models.

@AnthropicAI @code Great to see it

@AnthropicAI @code @grok Is it free to use

@AnthropicAI @code Yes, Teams is also a big update

@AnthropicAI @code Oh yeah!

@AnthropicAI @code Do we actually get the raw performance, like the claude code, or is it distilled in @GitHubCopilot?