Benchmark Almanac

MosaicML

We track 2 benchmarks that MosaicML made or helped make. Its models are on its lab page.

0 in use 2 with public code 1 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
Mosaic Eval Gauntlet v0.3.0 Models complete tasks across reading, reasoning, world knowledge, language understanding, and long context. MosaicML groups results by capability. Language reasoning 0.52 Not in cards
Mosaic Eval Gauntlet Original Models complete tasks in six capability groups. MosaicML compared the MPT models across these groups. Language reasoning Jun 2023 Not in cards