Miju Labs

← All buyers · Black Forest Labs

Axel Sauer

Co-founder & Research Scientist · Black Forest Labs

high confidenceTier A · Talk to firstImage, video & audio modelsimage

Sauer's own site lists him as Co-Founder & Research Scientist at Black Forest Labs, which upgrades the 'research staff' label in the index. He is the distillation and fast-inference specialist — StyleGAN-T, Adversarial Diffusion Distillation, Latent ADD — and a co-author on the SD3 rectified-flow paper that won a best-paper award at ICML 2024. Distillation work lives or dies on preference evaluation, because the standard question is whether the fast student still looks as good as the teacher.

Background

Sauer did his doctoral work on generative adversarial networks and text-to-image synthesis (StyleGAN-XL, StyleGAN-T with Nvidia collaborators), then joined Stability AI where he led Adversarial Diffusion Distillation — the method behind SDXL Turbo — and co-authored the SD3 paper. He followed Rombach, Blattmann and Esser to Black Forest Labs as a co-founder.

His research line is consistently about making high-quality generation fast: one-to-four-step sampling against the usual thirty to a hundred. That work is evaluated almost entirely by human preference, since automatic metrics do not capture the artefacts distillation introduces.

What they run now

Fast sampling and distillation inside the FLUX line — the Schnell and Klein variants, and whatever makes FLUX 3's video generation tractable. Notable papers: StyleGAN-T (ICML 2023 oral), Adversarial Diffusion Distillation (ECCV 2024 oral), Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation (2024), and Scaling Rectified Flow Transformers (ICML 2024, best paper).

What they buy

Inference. Distillation papers in this line report head-to-head human preference studies as their primary quality evidence, so Sauer is a recurring consumer of pairwise preference data on image and now video quality — specifically 'did the fast model lose anything' comparisons. He has not named a vendor publicly. A small, fast-turnaround expert preference panel is the shape of product he would actually use.

Career

  1. 2024 – presentCo-founder & Research Scientist, Black Forest Labs — Per his own site
  2. 2023 – 2024Research Scientist, Stability AI — Adversarial Diffusion Distillation; SD3
  3. c. 2019 – 2023PhD researcher, University of Tübingen / MPI-IS (doctoral research) — StyleGAN-XL, StyleGAN-T
  4. c. 2019–2023University of Tübingen / Max Planck Institute for Intelligent Systems, PhD, machine learning (generative models)

Interviews & talks

Why it matters here

High practical value and the most reachable BFL founder. He has an active X account, a personal site and an academic publication habit, which means he responds to technical artefacts. If Julian wants one BFL contact to start a real conversation with, Sauer is easier than Rombach and closer to the evaluation loop than Blattmann or Esser.

How to reach

Reach him on X (@AxSauer) or LinkedIn with a distillation-specific artefact: a small pairwise study of FLUX Schnell/Klein against the full model on dimensions distillation is known to damage — fine texture, hands, text rendering, motion coherence — rated by named professionals, with the rubric published. Speculation: framing it as 'your ADD papers use preference studies as the headline evidence; here is what a proper panel costs and looks like' is more likely to land than a services pitch.

What we could not establish
  • Exact PhD institution and completion date not confirmed from a primary source.
  • LinkedIn slug taken from his own site's link text and not independently verified.
  • Whether he personally commissions rater studies or hands that to others.

Why they are on the map

Co-author of FLUX.1 Kontext paper (June 2025) and known for adversarial diffusion distillation (LADD, which the post-training job also names); likely post-training/distillation lead. source — evidence dated 2025-06

Needs the data for a model; writes the spec.

How Black Forest Labs buys data

FLUX 3 launch (2026-07-23) reported head-to-head human preference results vs Luma Ray 3.2 (93%), Runway Gen-4.5 (77%) and Grok Imagine Video; raters and methodology not disclosed. Open role 'Data Procurement & Operations Lead' (Freiburg/SF, posted around Jan 2026) will own data vendors, specialist contractors, stock media and licensors, and build 'collection and evaluation systems through internal tools, vendor partnerships'. 'MTS - Post Training' role (2026) covers RLHF/DPO/reward modeling and 'human prefs'. FLUX.1 Kontext (2025-06) introduced crowd-sourced KontextBench.

Flat 'Member of Technical Staff' research org led by founders. Teams visible in hiring: Pretraining, Post Training, VLM, Image/Video Generation, Research Infra. Data procurement/ops is being set up as its own lead role, and no named holder was found.

Sources

  1. Data Procurement & Operations Lead — Black Forest Labs (Greenhouse)
  2. The AI startup that has quietly become one of Europe's most valuable AI companies (Fortune, 2026-02-17)
  3. FLUX 3 — Real World Models (BFL blog, 2026-07-23)
  4. Black Forest Labs on FLUX — a16z AI Podcast transcript (2024-08-18)
  5. Flux (text-to-image model) — Wikipedia
  6. Black Forest Labs: Europe's most-hyped — and elusive — startup? (Sifted, 2025-04-24)
  7. Axel Sauer — personal site
  8. dblp: Axel Sauer