Abundant launches SWE-Marathon coding benchmark, where Claude Opus 4.8 leads frontier models with a 26% resolution rate
The long-horizon tasks include building Slack from scratch.
Story Brief
The long-horizon tasks include building Slack from scratch.
Related Stories
4 linkedDAIR.AI AcademyLearn AI with hands-on courses on AI agents, RAG, LLMs, and more. Powered by AI tutoring and a supportive community.29d agodair.ai
SWE-Marathon20 multi-hour SWE tasks spanning library reproductions, full-stack product clones, and ML engineering. 1,300 logged trials; frontier configs stay below 19% task resolution.31d agoSWE-Marathon
SWE-Marathon20 multi-hour SWE tasks spanning library reproductions, full-stack product clones, and ML engineering. 1,300 logged trials; frontier configs stay below 19% task resolution.31d agoSWE-Marathon
SWE-Marathon20 multi-hour SWE tasks spanning library reproductions, full-stack product clones, and ML engineering. 1,300 logged trials; frontier configs stay below 19% task resolution.31d agoSWE-MarathonSWE-Marathon20 multi-hour SWE tasks spanning library reproductions, full-stack product clones, and ML engineering. 1,300 logged trials; frontier configs stay below 19% task resolution.31d agoSWE-Marathon
