Claude Sonnet 4.6

Out of the gate, Claude Sonnet 4.6 is already excelling at complex code fixes, especially when searching across large codebases is essential. For teams running agentic coding at scale, we're seeing strong resolution rates and the kind of consistency developers need.

Claude Sonnet 4.6 produced the best iOS code we've tested for Rakuten AI. Better spec compliance, better architecture, and it reached for modern tooling we didn't ask for, all in one shot. The results genuinely surprised us.

The performance-to-cost ratio of Claude Sonnet 4.6 is extraordinary — it’s hard to overstate how fast Claude models have been evolving in recent months. Sonnet 4.6 outperforms on our orchestration evals, handles our most complex agentic workloads, and keeps improving the higher you push the effort settings.

Claude Sonnet 4.6 is noticeably more capable on the hard problems where Sonnet 4.5 falls short, and shows strength on tasks that normally require the autonomy and agentic capabilities of an Opus model.

Claude Sonnet 4.6 punches way above its weight class for the vast majority of real-world PRs, and even improving more than 10 points on the hardest bug finding problems over Sonnet 4.5. For the code that actually ships, it's become our default.

On our long-horizon coding eval — where every feature builds on the last and you live with your earlier decisions — Sonnet 4.6 matches Opus 4.5. Fewer tokens. Faster. For most engineering, start here.

Claude Sonnet 4.6 has perfect design taste when building frontend pages and data reports, and it requires far less hand-holding to get there than anything we've tested before.

Claude Sonnet 4.6 is faster, cheaper, and more likely to nail things on the first try. That was a surprising set of improvements, and we didn't expect to see it at this price point.

Claude Sonnet 4.6 benchmarks near Opus-level on the coding tasks we care about, with meaningfully better instruction-following and tool reliability than its predecessor. We’re transitioning our Sonnet traffic over to this model.

Claude Sonnet 4.6 meaningfully improves the answer retrieval behind our core product — we saw a significant jump in answer match rate, compared to Sonnet 4.5, in our Financial Services Benchmark, with better recall on the specific workflows our customers depend on.

Claude Sonnet 4.6 rivals the best models on UI layout and leaps past the previous Sonnet. It's the model we want building apps for our users.

We're moving the majority of our traffic to Claude Sonnet 4.6. With adaptive thinking and high effort, we see Opus-level performance on all but our hardest analytical tasks with a more efficient and flexible profile. At Sonnet pricing, it's an easy call for our workloads.

Claude Sonnet 4.6 hit 94% on our complex insurance computer use benchmark, the highest of any Claude model we tested. It reasons through failures and self-corrects in ways we haven't seen before.

Box evaluated how Claude Sonnet 4.6 performs when tested on deep reasoning and complex agentic tasks across real enterprise documents. It demonstrated significant improvements, outperforming Claude Sonnet 4.5 in heavy reasoning Q&A by 15 percentage points.

Claude Sonnet 4.6 delivers frontier-level results on complex app builds and bugfixing. It's becoming our go-to for the kind of deep codebase work that used to require more expensive models.

Claude Sonnet 4.6's consistency across complex multi-character stories is genuinely impressive — it keeps each character's voice distinct in ways that make our narratives feel more alive.

We’ve been impressed by how accurately Claude Sonnet 4.6 handles complex computer use. It’s a clear improvement over anything else we’ve tested in our evals.

Claude Sonnet 4.6 excels at the behaviors that matter most for real knowledge work, scoring high in our internal evals, while making fewer tool calls and hitting fewer tool errors. The result is an AI that handles complicated, multi-step requests more reliably and produces more consistent, polished work.

Claude Sonnet 4.6 matches Opus 4.6 performance on OfficeQA, which measures how well a model can read enterprise documents (charts, PDFs, tables), pull the right facts, and reason from those facts. It's a meaningful upgrade for document comprehension workloads.

Claude Sonnet 4.6 achieves speed, quality, and economy. We especially like the 1M token context for larger projects. It surprises with intuitive and thoughtful comments, pushing back thoughtfully and genuinely understanding the goals of its work.

Claude Sonnet 4.6 is a notable improvement over Sonnet 4.5 across the board, including long horizon tasks and more difficult problems.

Claude Sonnet 4.6 has meaningfully closed the gap with Opus on bug detection, letting us run more reviewers in parallel, catch a wider variety of bugs, and do it all without increasing cost.

For the first time, Claude Sonnet 4.6 brings frontier-level reasoning in a smaller and more cost effective form factor. It provides a viable alternative if you are a heavy Opus user.

Claude Sonnet 4.6 is a significant leap forward on reasoning through difficult tasks. We find it especially strong on branched and multi-step tasks like contract routing, conditional template selection, and CRM coordination — exactly where our customers need strong model sense and reliability.

In Rovo Dev testing, Claude Sonnet 4.6 proved to be a highly effective main agent, leveraging subagents to sustain longer-running tasks.

Claude Sonnet 4.6 shows impressive progress in reasoning, code understanding, and memory—key ingredients for agentic automation. Our early tests integrating it into Postman Agent Mode point to smoother, more capable end-to-end workflows, and we’re excited about what this means for API developers building with automation at scale.

Claude Sonnet 4.6 was exceptionally responsive to direction — delivering precise figures and structured comparisons when asked, while also generating genuinely useful ideas on trial strategy and exhibit preparation.

In our zero-shot app building experiments, Sonnet 4.6 ran up to 3-4x longer without intervention, producing functional apps on par with the Opus series.

Claude Sonnet 4.6 produced zero hallucinated links in our computer use evals. Previously, we saw about one in three hallucinated links. That kind of reliability is what actually makes browser automation deployable at scale.

On our filesystem benchmark, Claude Sonnet 4.6 is 70% more token-efficient than Sonnet 4.5 model with a 38% accuracy improvement. That changes the economics of what we can build.