{"id":"6445c5719031","page":"/c/6445c5719031/","topic":"SWE-bench Verified accuracy for Claude models","publisher":"@EpochAIResearch","official":false,"platform":"X","publishedAt":"2025-05-28T19:46:04.000Z","model":"Claude 4","form":"dumbbell","metrics":{"views":32143,"likes":338,"reposts":31,"quotes":7,"replies":3,"saves":81},"readAt":"2026-10-12T21:30:08.589Z","chartsInPost":1,"score":61,"scoreWeights":{"views":0.4,"likes":0.3,"eng":0.15,"save":0.15},"quadrant":"loved","ranks":{"score":{"industry":{"n":442,"of":1162},"company":{"n":10,"of":59}},"views":{"industry":{"n":632,"of":1162},"company":{"n":14,"of":59}},"likes":{"industry":{"n":542,"of":1162},"company":{"n":13,"of":59}},"eng":{"industry":{"n":214,"of":1162},"company":{"n":38,"of":59}},"save":{"industry":{"n":63,"of":1162},"company":{"n":11,"of":59}}},"percentiles":{"industry":{"views":0.45654044750430295,"likes":0.5339931153184165,"eng":0.8162650602409639,"save":0.9462134251290878},"company":{"views":0.7711864406779662,"likes":0.788135593220339,"eng":0.3644067796610169,"save":0.8220338983050848}},"peers":{"industry":1162,"company":59,"companyName":"@epochairesearch"},"medians":{"industry":{"views":39945,"likes":289,"eng":0.007331788144862199,"save":0.0007642791742968587},"company":{"views":5278,"likes":63,"eng":0.013723883654240066,"save":0.0012807487454203996},"form":{"n":8,"views":15190,"eng":0.009268942846856849}},"ratios":{"vsCompanyViews":6.089996210685866,"vsCompanyLikes":5.365079365079365,"viewsPerFollower":null,"viewsPerDay":64.0206631176987},"followers":null,"reads":[{"at":"2026-10-12T15:32:45.338Z","views":32143,"likes":338,"reposts":31,"quotes":7,"replies":3,"saves":81,"source":"x-search-cache"},{"at":"2026-10-12T15:34:33.759Z","views":32143,"likes":338,"reposts":31,"quotes":7,"replies":3,"saves":81,"source":"x-search-cache"}],"chartData":"/api/charts/6445c5719031.json","copy":{"en":{"why":[{"factor":"reach-vs-company","text":"At 32.1K views, this post runs 6.1x the 5.3K median for @EpochAIResearch company posts."},{"factor":"saves","text":"Its 0.25% save rate sits in the 95th percentile among AI chart posts, against a 0.076% median."},{"factor":"launch-day","text":"It ranked #1 of 4 posts from this account on May 28, 2025."}],"verdict":"It lands in the Small but loved quadrant, at #442 of 1,162 AI chart posts.","news":{"headline":"@EpochAIResearch charts SWE-bench Verified accuracy for Claude models","dek":"A dumbbell chart on SWE-bench Verified accuracy for Claude models drew 32.1K views and a 0.25% save rate.","what":"@EpochAIResearch posted a dumbbell chart on SWE-bench Verified accuracy for Claude models on May 28, 2025. It came with 3 other posts from the account that day, on GPQA Diamond accuracy, SWE-bench Verified accuracy, and OTIS Mock AIME 2024-2025 accuracy.","numbers":"The post reached 32.1K views, ranking #632 of 1,162 AI chart posts and #14 of 59 @EpochAIResearch posts. Its score of 61 placed it at #442 of 1,162 AI chart posts.","lesson":"A 0.25% save rate, at the 95th percentile among AI chart posts, shows readers bookmark this dumbbell chart at a rate far above the 0.076% median. The 5 models shown sit in one chart, and that chart is what gets kept.","implication":"At #10 of 59 @EpochAIResearch posts by score, this chart sits well inside the account's own top tier. Its 32.1K views fall short of the 39.9K median among AI chart posts, so the save rate carries the result."}},"zh":{"why":[{"factor":"reach-vs-company","text":"浏览量3.2万，是@EpochAIResearch中位数的6.1倍，在59条帖子中排第14。"},{"factor":"saves","text":"收藏率0.25%，在1162条AI图表帖中排第63，处于第95百分位。"},{"factor":"launch-day","text":"发布当天4条帖子中，本条浏览量排第1。"}],"verdict":"范围小，但很受欢迎，在1162条AI图表帖中排第442。","news":{"headline":"@EpochAIResearch发布Claude模型对比图表","dek":"浏览量低于中位数，但收藏率极高。","what":"2025年5月28日，@EpochAIResearch发布1张图表，对比5个Claude模型。当天还有3条其他帖子。","numbers":"浏览量3.2万，在1162条AI图表帖中排第632，低于中位数4万。收藏率0.25%，高于中位数0.076%。","lesson":"收藏率排第63，前10%，但浏览量仅排第632，范围小却受欢迎。","implication":"对@EpochAIResearch而言，此帖浏览量是其中位数的6.1倍，但低于AI图表帖中位数。"}},"lane":"en:amd/DeepSeek-V4.1-Flash zh:amd/DeepSeek-V4-Flash","checkedAt":"2026-10-13T20:40:06.268Z"}}