{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/gaussian-mixture-flow-matching-models","title":"Gaussian Mixture Flow Matching Models","arxiv_id":"2504.05304","date":"2025-04-07","proceeding":null,"authors":["Hansheng Chen","Kai Zhang","Hao Tan","Zexiang Xu","Fujun Luan","Leonidas Guibas","Gordon Wetzstein","Sai Bi"],"abstract":"Diffusion models approximate the denoising distribution as a Gaussian and predict its mean, whereas flow matching models reparameterize the Gaussian mean as flow velocity. However, they underperform in few-step sampling due to discretization error and tend to produce over-saturated colors under classifier-free guidance (CFG). To address these limitations, we propose a novel Gaussian mixture flow matching (GMFlow) model: instead of predicting the mean, GMFlow predicts dynamic Gaussian mixture (GM) parameters to capture a multi-modal flow velocity distribution, which can be learned with a KL divergence loss. We demonstrate that GMFlow generalizes previous diffusion and flow matching models where a single Gaussian is learned with an $L_2$ denoising loss. For inference, we derive GM-SDE/ODE solvers that leverage analytic denoising distributions and velocity fields for precise few-step sampling. Furthermore, we introduce a novel probabilistic guidance scheme that mitigates the over-saturation issues of CFG and improves image generation quality. Extensive experiments demonstrate that GMFlow consistently outperforms flow matching baselines in generation quality, achieving a Precision of 0.942 with only 6 sampling steps on ImageNet 256$\\times$256.","url_abs":"https://arxiv.org/abs/2504.05304v2","url_pdf":"https://arxiv.org/pdf/2504.05304v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"gaussian-mixture-flow-matching-models","repo_url":"https://github.com/lakonik/gmflow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2504.05304","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.05304"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lakonik/gmflow","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":1,"unverified":5},"by_repo_kind":{"official":{"samples":6,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"7bcb35f9f94c22a5","entry":"probabilistic_guidance_jit","repo":"lakonik/gmflow","repo_kind":"official","path":"lib/models/diffusions/gmflow.py","file_url":"https://github.com/lakonik/gmflow/blob/HEAD/lib/models/diffusions/gmflow.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7bcb35f9f94c22a5"}},{"code_sha256_prefix":"b35d6e215023c884","entry":"denoising_gm_convert_to_mean_jit","repo":"lakonik/gmflow","repo_kind":"official","path":"lib/models/diffusions/gmflow.py","file_url":"https://github.com/lakonik/gmflow/blob/HEAD/lib/models/diffusions/gmflow.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b35d6e215023c884"}},{"code_sha256_prefix":"1103b2593a21779a","entry":"gm1d_inverse_cdf_init_jit","repo":"lakonik/gmflow","repo_kind":"official","path":"lib/ops/gmflow_ops/gmflow_ops.py","file_url":"https://github.com/lakonik/gmflow/blob/HEAD/lib/ops/gmflow_ops/gmflow_ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1103b2593a21779a"}},{"code_sha256_prefix":"8f5d2087a7aecdf4","entry":"gm1d_pdf_cdf","repo":"lakonik/gmflow","repo_kind":"official","path":"lib/ops/gmflow_ops/gmflow_ops.py","file_url":"https://github.com/lakonik/gmflow/blob/HEAD/lib/ops/gmflow_ops/gmflow_ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8f5d2087a7aecdf4"}},{"code_sha256_prefix":"cfa36349a79bf6ff","entry":"gm_kl_loss","repo":"lakonik/gmflow","repo_kind":"official","path":"train_toymodel.py","file_url":"https://github.com/lakonik/gmflow/blob/HEAD/train_toymodel.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cfa36349a79bf6ff"}},{"code_sha256_prefix":"2d7ebd2eed334968","entry":"psd_inverse","repo":"lakonik/gmflow","repo_kind":"official","path":"lib/ops/gmflow_ops/gmflow_ops.py","file_url":"https://github.com/lakonik/gmflow/blob/HEAD/lib/ops/gmflow_ops/gmflow_ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2d7ebd2eed334968"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}