The recent proliferation of smart devices necessitates methods to learn small-sized models. This paper demonstrates that if there are m features in total but only n = o(√m) features are required to distinguish examples, with 𝛺(log m) training examples and reasonable settings, it is possible to obtain a good model in a succinct representation using n log₂ ^m⁄_n + o(m) bits, by using a pipeline of existing compression methods: L1-regularized logistic regression, feature hashing, Elias–Fano indices, and randomized quantization. An experiment shows that a noun phrase chunking task for which an existing library requires 27 megabytes can be compressed to less than 13 kilobytes without notable loss of accuracy.