Warm-Start Flow Matching provides a guaranteed speedup for image/text generation by using lightweight models as initial priors.
March 23, 2026
Original Paper
Warm-Start Flow Matching for Guaranteed Fast Text/Image Generation
arXiv · 2603.19360
The Takeaway
Rather than starting from pure noise, WS-FM initializes the ODE from a 'draft' sample generated by a cheap model. This allows the expensive flow matching process to start at a much later time-step, drastically reducing the number of NFEs required for high-quality samples.
From the abstract
Current auto-regressive (AR) LLMs, diffusion-based text/image generative models, and recent flow matching (FM) algorithms are capable of generating premium quality text/image samples. However, the inference or sample generation in these models is often very time-consuming and computationally demanding, mainly due to large numbers of function evaluations corresponding to the lengths of tokens or the numbers of diffusion steps. This also necessitates heavy GPU resources, time, and electricity. In