{"tasks":[{"slug":"c6-mode-choice-canada","scope":"international","geography":"Montreal–Toronto, Canada","phase":4,"family":"individual","scores":[{"slug":"llama","value":0.702258,"skill":-15.474471758612207,"n_valid":200,"name":"Llama-3.3-70B (API, temp 0)","display":"0.702","protocol":"Llama-3.3-70B (API, temp 0)","url":"/tasks/c6-mode-choice-canada"},{"slug":"gptoss","value":0.708658,"skill":-16.52684370632247,"n_valid":200,"name":"gpt-oss-120B (API, temp 0)","display":"0.709","protocol":"gpt-oss-120B (API, temp 0)","url":"/tasks/c6-mode-choice-canada"},{"slug":"gemma","value":0.71012,"skill":-16.76724492312751,"n_valid":200,"name":"Gemma-4-26B (API, temp 0)","display":"0.710","protocol":"Gemma-4-26B (API, temp 0)","url":"/tasks/c6-mode-choice-canada"}],"n_items":200,"n_test":200,"metric":"Brier (sum over modes)","higher":false,"unit":"","baseline":0.60815,"baseline_display":"0.608","source":"ModeCanada, distributed with the R package mlogit","url":"/tasks/c6-mode-choice-canada","version":"historical","human":null,"best":{"slug":"llama","value":0.702258,"skill":-15.474471758612207,"n_valid":200,"name":"Llama-3.3-70B (API, temp 0)","display":"0.702","protocol":"Llama-3.3-70B (API, temp 0)","url":"/tasks/c6-mode-choice-canada"},"leaders":[{"slug":"llama","value":0.702258,"skill":-15.474471758612207,"n_valid":200,"name":"Llama-3.3-70B (API, temp 0)","display":"0.702","protocol":"Llama-3.3-70B (API, temp 0)","url":"/tasks/c6-mode-choice-canada"}],"status":"evaluated","title":"Predict individual mode choice in the Montreal-Toronto corridor"}],"aggregation":"Per task and version only; no cross-task average"}