Claude Sonnet 5.5 Is Live on Multos — Anthropic's Fastest Sonnet Yet
Anthropic released Claude Sonnet 5.5 on September 28, 2026. It runs 30%+ faster than Sonnet 5, scores 70.6% on Terminal-Bench 4.0 versus Sonnet 5's 10.3%, and costs up to 30% less per task. It's live on Multos Pro and above, in the same bracket as Sonnet 5.
Anthropic shipped Claude Sonnet 5.5 on September 28, 2026, and it is available on Multos as of today. It is the second model in the Claude 5.5 family, positioned as the fast, lower-cost complement to Opus 5.5. The short version: it is dramatically better than Sonnet 5 at agentic coding, generates output 30%+ faster, and typically costs less per task because it needs fewer tokens to finish the same work. It sits on Multos Pro and above, in the same bracket as Sonnet 5 — so if you already had Sonnet 5, you now have a strictly better option at the same access level.
The Benchmark That Matters
The headline number is Terminal-Bench 4.0, an agentic coding evaluation that measures how well a model completes complex, multi-step professional tasks inside a command-line interface. Sonnet 5.5 scores 70.6%. Sonnet 5 scored 10.3%. That is not a marginal improvement. It is the difference between a model that can hold a multi-step terminal task together and one that cannot. For anyone building on Multos, where agents routinely read files, run commands, and edit code across many turns, that gap is the whole game. The rest of the benchmark sheet is consistent. On CursorBench 4.0, which tests models on ambiguous, multi-file tasks taken from real Cursor sessions, Sonnet 5.5 scores 55.5% — within about two points of Opus 5.5. On FrontierCode v1.1 it scores 46.2% at Max effort. On GDPval-AA v2.1, a test of real-world knowledge work across occupations, it lands at 1844 Elo, two points below Opus 5.5. On Humanity's Last Exam with tools, 64.5%. On OSWorld 2.1 computer use, 80.1%. On Chartography visual chart recognition, 61.6% with no tools — against Sonnet 5's 15.6%.
Speed and Cost: The Compounding Win
Sonnet 5.5 generates output 30%+ faster than Sonnet 5, making it Anthropic's fastest Sonnet model to date. That alone changes how a session feels — less waiting between idea and result. But the more important number is tokens per task. Anthropic's testing found Sonnet 5.5 costs up to 30% less per task than its predecessor, not because the per-token price dropped, but because the model needs far fewer tokens to do the same work. Early testers reported it batches tool calls together more than Sonnet 5, leading to fewer steps and lower costs. Lovable's team measured a third fewer tool calls and roughly half the shell runs to finish a task. On Multos, where your monthly pool is measured in tokens, fewer tokens per task means your pool goes further. A model that solves the same problem in fewer steps is cheaper in practice even at an identical per-token rate.
What This Means for Building on Multos
Sonnet 5.5 is the everyday workhorse. Anthropic positions it as strongest at well-scoped tasks, bug fixing, and producing polished documents, slides, and spreadsheets — with a sharp eye for design. That maps directly onto the most common Multos workflows: fixing a bug, refactoring a module, shipping a feature, generating a report. Because it is wired into Multos auto-routing and the fallback chain, you do not have to know when to pick it. If you are on auto and Sonnet 5.5 is the right fit for the task, the router reaches for it. If it is ever unavailable, your request falls through to a capable alternative instead of failing. For coding specifically, the jump is the point. Sonnet 5.5 understands a codebase quickly and works in fewer, more robust steps. In a Multos session that spans many turns — read the repo, find the bug, write the fix, explain the change — that efficiency compounds. You get to working code in fewer iterations, and you spend less of your pool getting there.
Safety and Availability
Sonnet 5.5 is the first Sonnet model to launch with cyber safeguards and fallbacks like those Anthropic developed for its most capable models, because its cybersecurity capabilities are comparable to Opus 5's. Routine software development is unaffected — you can still find and fix bugs in your own code — but higher-risk cybersecurity tasks visibly fall back to Sonnet 5. It is also the first Sonnet model to ship with safety classifiers that prevent reasoning extraction, and it expands preserved thinking so Claude's reasoning cannot be decoupled from the account that created it. It is available with zero data retention, and on all major platforms including AWS, Google Cloud, and Microsoft Azure. On Multos, that means you get it through the managed connection — no API key, no provider setup, no separate billing.
Where to Find It
Claude Sonnet 5.5 is available on Multos Pro and above, alongside Sonnet 5 in the same bracket. Find it in the model dropdown under the Anthropic group, or let Auto Router route to it for well-scoped coding and knowledge-work tasks where it is the right cost-performance fit. If you were already using Sonnet 5, there is no reason to stay on it. Sonnet 5.5 is faster, materially stronger on agentic coding, and cheaper per task. Same tier, same access, better model.
Ready to Get Started?
Claude Sonnet 5.5 is live on Multos Pro and above. Anthropic's fastest Sonnet yet — 70.6% on Terminal-Bench 4.0, 30%+ faster, and up to 30% cheaper per task. Select it from the model dropdown or let Auto Router pick it for your everyday coding work.
Start Building FreeRelated Articles
Claude Opus 5.5 Is Live on Multos — Fable-Class Work at 40% Lower Cost
Anthropic released Claude Opus 5.5 today and it's immediately available on Multos Elite. It matches Claude Fable 5.1 on most tasks, costs 40% less to run than Opus 5, and is 30% faster. On Terminal-Bench 4.0 it scores 66.4% — higher than GPT-6 Astra, higher than Opus 5, higher than Fable 5.1. Here's what that actually means for building apps and games on Multos.
Read more →GPT-6 Sol and GPT-6 Luna Are Live — OpenAI Halves the Price of Its Best Models
OpenAI released GPT-6 Sol and GPT-6 Luna today, cutting prices by 50% versus GPT-5.6. Sol drops from $4/$20 to $2/$10 per million tokens. Luna drops from $0.20/$1.20 to $0.10/$0.50. Intelligence Index scores hold steady, hallucinations drop significantly, and both models are live on Multos now — Sol on Pro, Luna on Lite.
Read more →MiMo V2.6 Flash and Pro Are Now on Multos — Frontier-Grade Agents at Flash-Tier Pricing
Xiaomi released MiMo V2.6 Pro and Flash on September 21, 2026. Pro scored 46 on the Artificial Analysis Intelligence Index — the highest any open-weight model has ever scored, at $0.13 per task. Flash costs $0.14/$0.28 per million tokens and trails Pro by only 4 points on DeepSWE. Both are live on Multos Lite via DeepInfra.
Read more →