{"tasks":[{"slug":"c6-rank-dresden-traffic","scope":"germany","geography":"Dresden, Germany","phase":2,"family":"traffic","scores":[{"slug":"gpt5","value":58.333333333333336,"skill":37.500000000000014,"n_valid":12,"name":"GPT-5 (no search)","display":"58.333%","protocol":"GPT-5 (no search)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"deepseek","value":50.0,"skill":25.000000000000007,"n_valid":12,"name":"DeepSeek-V4-Flash (API, temp 1.0, majority of 6 usable runs)","display":"50.000%","protocol":"DeepSeek-V4-Flash (API, temp 1.0, majority of 6 usable runs)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"gemini","value":50.0,"skill":25.000000000000007,"n_valid":12,"name":"Gemini (no search)","display":"50.000%","protocol":"Gemini (no search)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"glm53","value":41.666666666666664,"skill":12.500000000000004,"n_valid":12,"name":"GLM-5.3 (API, temp 1.0, majority of 3 usable runs)","display":"41.667%","protocol":"GLM-5.3 (API, temp 1.0, majority of 3 usable runs)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"minimax3","value":41.666666666666664,"skill":12.500000000000004,"n_valid":12,"name":"MiniMax-M3 (API, temp 1.0, majority of 10 runs)","display":"41.667%","protocol":"MiniMax-M3 (API, temp 1.0, majority of 10 runs)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"kimi","value":33.333333333333336,"skill":1.0658141036401503e-14,"n_valid":12,"name":"Kimi-K2 (no search)","display":"33.333%","protocol":"Kimi-K2 (no search)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"qwen","value":33.333333333333336,"skill":1.0658141036401503e-14,"n_valid":12,"name":"Qwen3.8-27B (API, temp 1.0, majority of 9 usable runs)","display":"33.333%","protocol":"Qwen3.8-27B (API, temp 1.0, majority of 9 usable runs)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"gemma","value":25.0,"skill":-12.499999999999991,"n_valid":12,"name":"Gemma-4-26B (API, temp 1, majority of 5)","display":"25.000%","protocol":"Gemma-4-26B (API, temp 1, majority of 5)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"gptoss","value":25.0,"skill":-12.499999999999991,"n_valid":12,"name":"gpt-oss-120B (API, temp 1, majority of 5)","display":"25.000%","protocol":"gpt-oss-120B (API, temp 1, majority of 5)","url":"/tasks/c6-rank-dresden-traffic"},{"slug":"llama","value":0.0,"skill":-49.99999999999999,"n_valid":12,"name":"Llama-3.3-70B (API, temp 1, majority of 5)","display":"0.000%","protocol":"Llama-3.3-70B (API, temp 1, majority of 5)","url":"/tasks/c6-rank-dresden-traffic"}],"n_items":12,"n_test":12,"metric":"Accuracy","higher":true,"unit":"%","baseline":33.33333333333333,"baseline_display":"33.333%","source":"Landeshauptstadt Dresden · official traffic counts","url":"/tasks/c6-rank-dresden-traffic","version":"historical","human":null,"best":{"slug":"gpt5","value":58.333333333333336,"skill":37.500000000000014,"n_valid":12,"name":"GPT-5 (no search)","display":"58.333%","protocol":"GPT-5 (no search)","url":"/tasks/c6-rank-dresden-traffic"},"leaders":[{"slug":"gpt5","value":58.333333333333336,"skill":37.500000000000014,"n_valid":12,"name":"GPT-5 (no search)","display":"58.333%","protocol":"GPT-5 (no search)","url":"/tasks/c6-rank-dresden-traffic"}],"status":"evaluated","title":"Rank Dresden streets by daily traffic volume"}],"aggregation":"Per task and version only; no cross-task average"}