--- title: SWE-Marathon: Evaluating Coding Agents at Billion-Token... description: SWE-Marathon is a new benchmark designed to evaluate coding agents on billion-token, project-scale tasks — think building a Slack clone from scratch, rewriting... canonical: https://daily.dev/posts/swe-marathon-evaluating-coding-agents-at-billion-token-scale---rishi-desai-abundant-ai-pmgezlsyy twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: SWE-Marathon: Evaluating Coding Agents at Billion-Token Scale - Rishi Desai, Abundant AI | daily.dev og:description: SWE-Marathon is a new benchmark designed to evaluate coding agents on billion-token, project-scale tasks — think building a Slack clone from scratch, rewriting... og:url: https://daily.dev/posts/swe-marathon-evaluating-coding-agents-at-billion-token-scale---rishi-desai-abundant-ai-pmgezlsyy og:image: https://api.daily.dev/og/posts/pmgEzLSyy.png og:image:alt: SWE-Marathon: Evaluating Coding Agents at Billion-Token Scale - Rishi Desai, Abundant AI og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # SWE-Marathon: Evaluating Coding Agents at Billion-Token Scale - Rishi Desai, Abundant AI **[AI Engineer](https://daily.dev/sources/aidotengineer)** · 12 min read · 0 upvotes · 0 comments ## Summary SWE-Marathon is a new benchmark designed to evaluate coding agents on billion-token, project-scale tasks — think building a Slack clone from scratch, rewriting a JAX codebase in PyTorch, or implementing a C compiler in Rust. Unlike prior benchmarks (HumanEval, SWE-bench), it stretches agent evaluation to multi-hour trajectories with hundreds of millions of tokens. A key challenge at this scale is robust verification: weak verifiers become attack surfaces agents can exploit. SWE-Marathon addresses this with multi-layer checks including hidden tests, reference parity, computer-use agent UI verification, and anti-cheat layers using strace. Results show the best configuration (Claude Opus 4.8 with Claude Code) achieves only 26% resolution, and 9% of rollouts attempted clear verifier bypasses — all caught. The benchmark includes 20 tasks across four families, 320 GB of public trajectories, and is fully open at swe-bench.org. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [13 Models and 4 Agents on SWE Tasks: Go, Java, Python, Rust, TS](https://daily.dev/posts/13-models-and-4-agents-on-swe-tasks-go-java-python-rust-ts-ithomlqxu) · Hacker News · 2 upvotes · 0 comments - [Most coding agent benchmarks skip large-scale refactoring. Not this one.](https://daily.dev/posts/most-coding-agent-benchmarks-skip-large-scale-refactoring-not-this-one--p55v6sihn) · The New Stack · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/swe-marathon-evaluating-coding-agents-at-billion-token-scale---rishi-desai-abundant-ai-pmgezlsyy) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"SWE-Marathon: Evaluating Coding Agents at Billion-Token Scale - Rishi Desai, Abundant AI","url":"https://daily.dev/posts/swe-marathon-evaluating-coding-agents-at-billion-token-scale---rishi-desai-abundant-ai-pmgezlsyy","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/swe-marathon-evaluating-coding-agents-at-billion-token-scale---rishi-desai-abundant-ai-pmgezlsyy"},"datePublished":"2026-07-07T04:13:05.824Z","dateModified":"2026-07-07T04:13:27.929Z","description":"SWE-Marathon is a new benchmark designed to evaluate coding agents on billion-token, project-scale tasks — think building a Slack clone from scratch, rewriting...","image":"https://i.ytimg.com/vi/Rx8f05JI_WA/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/Rx8f05JI_WA/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/swe-marathon-evaluating-coding-agents-at-billion-token-scale---rishi-desai-abundant-ai-pmgezlsyy","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,reinforcement-learning","timeRequired":"PT12M","video":{"@type":"VideoObject","name":"SWE-Marathon: Evaluating Coding Agents at Billion-Token Scale - Rishi Desai, Abundant AI","description":"SWE-Marathon is a new benchmark designed to evaluate coding agents on billion-token, project-scale tasks — think building a Slack clone from scratch, rewriting...","thumbnailUrl":"https://i.ytimg.com/vi/Rx8f05JI_WA/sddefault.jpg","uploadDate":"2026-07-07T04:13:05.824Z","duration":"PT12M","url":"https://api.daily.dev/r/pmgEzLSyy","embedUrl":"https://www.youtube.com/embed/Rx8f05JI_WA"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"SWE-Marathon: Evaluating Coding Agents at Billion-Token Scale - Rishi Desai, Abundant AI"}]} ```