In case you missed it, earlier this week we fixed one of the most common frustrations on https://t.co/8aEVBp4ZOc: hitting context limits mid-conversation. Claude now intelligently compacts earlier context automatically when you're nearing the limit so the chat can keep going.

We're still learning more about the model ourselves every day! Share any tips you have below.

If you want to quickly incorporate all these changes and migrate your app to Opus 4.5, use this migration Claude Code plugin we made https://t.co/V8ZdOilmfX

5. Vision is improved. Opus is better at image processing and data extraction, especially with multiple images. For dense images, give it a cropping tool to "zoom" in on regions. We've seen consistent uplift on image evals with this.

4. Opus 4.5 can be conservative exploring code. If the model proposes solutions without reading the code first, tell it directly: "ALWAYS read and understand relevant files before proposing edits. Do not speculate about code you have not inspected."

3. Opus 4.5 can at times overengineer and add extra files, abstractions, etc. To fix for your use case: add explicit prompting like "Only make changes that are directly requested. Keep solutions simple and focused."

2. Tool triggering rates may change. Opus 4.5 is more responsive to system prompts, so if your old prompts used aggressive language to reduce undertriggering, you may now see overtriggering. Dial back "CRITICAL: You MUST use this tool" to just "Use this tool when..."

1. The new effort parameter is powerful because it controls approximately how many Claude will use for an output. You can trade off intelligence for cost/latency with a single dial. Works on all tokens including thinking, responses, and tool calls.

We put together a prompting guide for Claude Opus 4.5 based on extensive internal testing by our research and applied AI teams. Here's what we've learned so far about getting the best results:

https://t.co/PutuHlGYre

fyi we made Claude for Excel is now live for all Max, Team, and Enterprise users. Opus 4.5 makes it meaningfully better at complex spreadsheet tasks.

Read the full story in our model card:

We had to remove the τ2-bench airline eval from our benchmarks table because Opus 4.5 broke it by being too clever. The benchmark simulates an airline customer service agent. In one test case, a distressed customer calls in wanting to change their flight, but they have a basic

All three features are available now in beta. Read the full blog for more details:

Tool Use Examples JSON Schema defines what's valid, not what's correct. Now you can show Claude concrete usage patterns directly in tool definitions to improve Claude's accuracy and knowledge when using tools.

Programmatic Tool Calling Claude orchestrates tools through code instead of individual round-trips. It writes Python, processes outputs in a sandbox, and controls what enters context.

Tool Search Tool Instead of loading all tool definitions upfront, Claude discovers tools on-demand. Mark tools with defer_loading: true and only pays tokens for tools Claude actually needs. Up to an 85% token reduction and big boost in accuracy on our MCP evals (79.5% to 88.1%)

Alongside the model, today we're launching three very useful API features for building agents that scale to hundreds of tools without context bloat. - Tool Search Tool - Programmatic Tool Calling - Tool Use Examples Here's how they work:

>Opus 4.5 "seems to be able to vibe code forever" I've found this to be very true. Much more to come here but basically you can set-and-forget this model as it works on coding task for you in the background. Feels like we hit a step change.

Pricing is $5/$25 per million tokens. Available now on the Claude API and all three major cloud platforms (Amazon Bedrock, Google Cloud's Vertex AI, and Microsoft Foundry). Read more here: