SWE-Marathon
A long-session benchmark for coding agents, designed to evaluate an AI system's ability to sustain continuous execution across software engineering tasks over extended interaction sessions.
Core Facts
Timeline (last 90 days)
DeepSWE focuses on complex, deeply dependent engineering problems; SWE-Marathon emphasizes long-chain, multi-stage tasks, simulating scenarios of sustained work over long periods.
In the DeepSWE 1.1 benchmark, Muse Wonder Full achieved 79.0%, exceeding Claude Opus 5.5's 74.2% and GPT-6 Astra's 74.1%.
Meta uses FocusBench to evaluate focus and SWE-Marathon to test sustained execution on long-horizon software projects.
SWE-Marathon is an end-to-end outcome-based evaluation that gives the agent a software project spanning multiple hours and observes whether it can stay on course without drifting over a long horizon.
Meta's chart shows Muse Wonder beating Claude Opus 5.5 on long-horizon coding tasks: DeepSWE score 78.8 versus 74.2, SWE-Marathon score 53.1 versus 46.3.
SWE-Marathon stretches tasks into dozens or even hundreds of steps of continuous editing, testing the agent's ability not to get lost in a long context.
Meta notes in the model card that SWE-Marathon is not a controlled like-for-like comparison.
On DeepSWE 1.1, Muse Wonder beats Claude Opus 5.5 78.8 to 74.2, leads 10.0 to 4.2 on the Harvey legal agent benchmark LAB-AA, and leads 53.1 to 46.3 on SWE-Marathon.
Muse Wonder Full holds a leading position in long-horizon coding benchmarks such as DeepSuite.
TomasuLLM achieves a 1.27× matched-progress improvement across 18 sessions on SWE-Marathon.
All Facts (10)
Meta's chart shows Muse Wonder beating Claude Opus 5.5 on long-horizon coding tasks: DeepSWE score 78.8 versus 74.2, SWE-Marathon score 53.1 versus 46.3.
65%UnverifiedDeepSWE focuses on complex, deeply dependent engineering problems; SWE-Marathon emphasizes long-chain, multi-stage tasks, simulating scenarios of sustained work over long periods.
50%UnverifiedIn the DeepSWE 1.1 benchmark, Muse Wonder Full achieved 79.0%, exceeding Claude Opus 5.5's 74.2% and GPT-6 Astra's 74.1%.
50%UnverifiedSWE-Marathon is an end-to-end outcome-based evaluation that gives the agent a software project spanning multiple hours and observes whether it can stay on course without drifting over a long horizon.
50%UnverifiedMeta uses FocusBench to evaluate focus and SWE-Marathon to test sustained execution on long-horizon software projects.
50%UnverifiedMeta notes in the model card that SWE-Marathon is not a controlled like-for-like comparison.
50%UnverifiedOn DeepSWE 1.1, Muse Wonder beats Claude Opus 5.5 78.8 to 74.2, leads 10.0 to 4.2 on the Harvey legal agent benchmark LAB-AA, and leads 53.1 to 46.3 on SWE-Marathon.
50%UnverifiedSWE-Marathon stretches tasks into dozens or even hundreds of steps of continuous editing, testing the agent's ability not to get lost in a long context.
50%UnverifiedMuse Wonder Full holds a leading position in long-horizon coding benchmarks such as DeepSuite.
50%UnverifiedTomasuLLM achieves a 1.27× matched-progress improvement across 18 sessions on SWE-Marathon.
50%