{"slug": "google-deepmind-s-new-benchmark-tests-agents-on-blueprints-schematics-and-decks", "title": "Google DeepMind's new benchmark tests agents on blueprints, schematics and polished decks", "summary": "Google DeepMind launched the Multimodal Knowledge Agent benchmark (MKA) on November 10th, a 200-task test built with Artificial Analysis to grade how AI agents handle visually dense professional files such as blueprints, schematics and polished decks. Google says its Gemini 4 Carbon model ranks second on quality while completing tasks at six times lower cost on average than the top-scoring model, a claim the company frames as making price per completed task a metric to compare alongside peak score. The benchmark was announced by Google DeepMind senior product manager Rohan Doshi and research director Ben Caine.", "body_md": "# Google DeepMind's new benchmark tests agents on blueprints, schematics and polished decks\n\n**Built with Artificial Analysis, MKA uses 200 expert-designed tasks to grade visual knowledge work. Google says Gemini 4 Carbon ranks second on quality at a lower cost.**\n\n        By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)\n        · Published \n        · Updated \n\nPrimary source: [Google Gemini (official)](https://blog.google/innovation-and-ai/models-and-research/gemini-models/multimodal-knowledge-agent-benchmark/)\n\n## Why it matters\n\nMKA evaluates agents on visual files and finished deliverables that define professional work. For model buyers, Google's second-place quality ranking paired with a claimed sixfold cost advantage makes price per completed task a number to compare alongside peak score.\n\nGoogle DeepMind senior product manager Rohan Doshi and research director Ben Caine launched the Multimodal Knowledge Agent benchmark on November 10th, a 200-task test built with Artificial Analysis to grade how AI agents handle visually dense professional files. [Google](https://blog.google/innovation-and-ai/models-and-research/gemini-models/multimodal-knowledge-agent-benchmark/?ref=runtimewire) says [Gemini 4 Carbon](https://runtimewire.com/models/google/gemini-4-carbon) ranks second on quality in the test while completing tasks at six times lower cost on average than the top-scoring model.…\n\nWhen you purchase through links in our articles, [we may earn a small commission](https://runtimewire.com/affiliate-monetization-standards). This doesn’t affect our editorial independence.", "url": "https://wpnews.pro/news/google-deepmind-s-new-benchmark-tests-agents-on-blueprints-schematics-and-decks", "canonical_source": "https://runtimewire.com/article/google-deepmind-mka-benchmark-visual-knowledge-work", "published_at": "2026-11-10 17:07:48+00:00", "updated_at": "2026-11-10 20:22:01.419621+00:00", "lang": "en", "topics": ["ai-agents", "artificial-intelligence", "ai-research", "large-language-models"], "entities": ["Google DeepMind", "Google", "Gemini 4 Carbon", "Artificial Analysis", "Multimodal Knowledge Agent benchmark", "Rohan Doshi", "Ben Caine"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/google-deepmind-s-new-benchmark-tests-agents-on-blueprints-schematics-and-decks", "markdown": "https://wpnews.pro/news/google-deepmind-s-new-benchmark-tests-agents-on-blueprints-schematics-and-decks.md", "text": "https://wpnews.pro/news/google-deepmind-s-new-benchmark-tests-agents-on-blueprints-schematics-and-decks.txt", "jsonld": "https://wpnews.pro/news/google-deepmind-s-new-benchmark-tests-agents-on-blueprints-schematics-and-decks.jsonld"}}