{"slug": "tomasullm-out-of-order-speculative-execution-for-llm-agents", "title": "TomasuLLM: Out-of-Order Speculative Execution for LLM Agents", "summary": "Researchers released TomasuLLM, a runtime that executes LLM agent tool calls out of trajectory order while preserving task-execution correctness, according to arXiv paper 2609.38201v1. TomasuLLM drafts future actions, runs them in isolated copy-on-write sandboxes, traces their dependencies and effects, and commits results in trajectory order only after validation against committed state, delivering 1.31x on 100 SWE-bench Verified tasks, 1.35x on 28 Terminal-Bench 2.0 tasks, and 1.27x matched progress on 18 SWE-Marathon sessions. Across 4,010 audited commit-validation records, the runtime produced zero false accepts.", "body_md": "arXiv:2609.38201v1 Announce Type: new \nAbstract: Long-running tools can dominate coding-agent latency: compilers, test suites, and repository commands take seconds to minutes while the agent idles. This observation stall presents the same tension that drove out-of-order processors -- asequential interface hides work that can be predicted and started early, but a speculative result may become visible only after it and every earlier step have been validated.\n  We present TomasuLLM, a runtime that executes agent tool calls out of trajectory order while preserving task-execution correctness. It drafts future actions, runs them in isolated copy-on-write sandboxes, traces their dependencies and effects, and commits results in trajectory order only after validation against committed state. Across three benchmarks spanning sub-second to minutes-long tool calls, TomasuLLM improves the reported benchmark means and scales with tool latency: 1.31x on 100 SWE-bench Verified tasks, 1.35x on 28 Terminal-Bench 2.0 tasks, and 1.27x matched progress on 18 SWE-Marathon sessions. Across 4,010 audited commit-validation records, it produces zero false accepts.", "url": "https://wpnews.pro/news/tomasullm-out-of-order-speculative-execution-for-llm-agents", "canonical_source": "https://arxiv.org/abs/2609.38201", "published_at": "2026-10-01 04:00:00+00:00", "updated_at": "2026-10-01 04:18:58.544237+00:00", "lang": "en", "topics": ["ai-agents", "large-language-models", "ai-research", "developer-tools"], "entities": ["TomasuLLM", "SWE-bench Verified", "Terminal-Bench 2.0", "SWE-Marathon", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/tomasullm-out-of-order-speculative-execution-for-llm-agents", "markdown": "https://wpnews.pro/news/tomasullm-out-of-order-speculative-execution-for-llm-agents.md", "text": "https://wpnews.pro/news/tomasullm-out-of-order-speculative-execution-for-llm-agents.txt", "jsonld": "https://wpnews.pro/news/tomasullm-out-of-order-speculative-execution-for-llm-agents.jsonld"}}