Muse Spark 1.1: benchmarks, price and rank

Its headline results

The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.

• APEX-Agents31.8%

  Long tasks written by investment banking analysts, management consultants and corporate lawyers from their own work, done across files and apps in a realistic work setup.

  50th of 56. Independent result from Epoch AI.

• MCP Atlas88.1%

  Multi-step tasks where the model has to pick and combine many real tools, connected through MCP, the standard way agents plug into apps and data.

  2nd of 33. The maker's own number, from Meta, via LLM Stats.

• SWE-Bench Pro61.5%

  Long changes to real codebases in several languages, often across many files, including code from private and copyleft projects that models are unlikely to have seen.

  18th of 50. The maker's own number, from Meta, via LLM Stats.

• Humanity's Last Exam62.1%

  About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.

  5th of 74. The maker's own number, from Meta, via LLM Stats.

• Vending-Bench 2$6,520

  The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.

  21st of 60. Independent result from Epoch AI.

• OSWorld-Verified80.8%

  Ordinary tasks on a real desktop computer using normal apps: a browser, a spreadsheet, an email client, a photo editor, the file manager.

  7th of 25. The maker's own number, from Meta, via LLM Stats.

• Toolathlon75.6%

  Long tasks across many real apps and their tools, such as email, calendars, spreadsheets, code repositories and online stores.

  2nd of 37. The maker's own number, from Meta, via LLM Stats.

• DeepSWE 1.153%

  Software engineering tasks run in the same simple agent setup for every model.

  Independent result from LLM Stats.

• SimpleQA Verified57.8%

  Short factual questions with one checked answer, asked without web search.

  21st of 78. Independent result from Epoch AI.

• CharXiv Reasoning88.4%

  Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.

  11th of 45. The maker's own number, from Meta, via LLM Stats.