Miju Labs

← All buyers · Black Forest Labs

A

Andreas Blattmann

Co-founder · Black Forest Labs

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

Blattmann is one of BFL's three founding researchers and the one most associated with its video line — he co-authored Stable Video Diffusion before founding the company. He speaks for BFL's strategy publicly (the open-weights-as-funnel argument) and is a plausible internal owner of video data quality, which is the modality where BFL's evaluation claims are weakest.

Background

Blattmann was a PhD researcher with Robin Rombach in Björn Ommer's group at LMU Munich and a co-author on the Latent Diffusion / Stable Diffusion papers. He moved to Stability AI, worked on Stable Video Diffusion and SD3, and co-founded Black Forest Labs in 2024.

In Fortune's February 2026 profile he set out BFL's commercial logic: open-weight releases act as an adoption funnel, while production use by enterprises requires a paid commercial licence. He has said he wants BFL to become the standard for visual intelligence while staying structurally a small lab.

What they run now

Founding researcher across BFL's image and video stack, now inside FLUX 3's joint image/video/audio model. Publicly he owns the open-weights strategy: Schnell/Klein under permissive licences to drive adoption, Dev under non-commercial terms, Pro behind the API. On the a16z podcast he described using classical computer-vision techniques to filter the worst parts of video training data — a hint that BFL's video curation has historically been automated rather than human-judged.

What they buy

Inference. As a founder he approves rather than originates data spend, but his stated reliance on classical CV filters for video suggests an unmet need for human judgement on video quality — motion artifacts, temporal consistency, shot grammar — which is exactly what Luma's published Ray3 rubric covers and BFL's FLUX 3 post does not. Likely buyer of expert video preference pairs and video eval rubrics once FLUX 3 video exits 'preliminary'.

Career

  1. 2024 – presentCo-founder, Black Forest Labs
  2. 2022 – 2024Researcher (Stable Video Diffusion, SD3), Stability AI
  3. 2019 – 2022PhD researcher, LMU Munich (Ommer group) — Latent Diffusion co-author
  4. 2019–2022Ludwig Maximilian University of Munich, PhD-track research, computer vision

On the record

open weights as a funnel · 2026-02

Frames open-weight releases as an adoption funnel that converts into paid enterprise licences. source

video data filtering · 2024-08

Says they used fairly classical computer-vision techniques to filter out the worst parts of video data. source

staying small · 2026-02

Wants BFL to become the standard for visual intelligence while keeping a small-lab structure. source

Interviews & talks

Why it matters here

Second-line at BFL but a genuine door: he is the founder who talks most openly about data pipelines, and video is BFL's newest and least-evaluated modality. If Julian's wedge is expert video judgement rather than image aesthetics, Blattmann is the founder whose problem it solves.

How to reach

Speculation: approach with a video-specific artefact, not a general pitch — e.g. a short annotated teardown of FLUX 3 video failure modes on the dimensions Luma publishes (motion artifacts, temporal consistency, physics) scored by named film/VFX professionals. German-language outreach is available and probably welcome. He has no public X handle found, so LinkedIn or a warm intro through the Freiburg/Munich diffusion-research network is the route.

What we could not establish
  • No confirmed LinkedIn or X handle found.
  • Whether he personally owns video data quality inside BFL, or whether that sits with Axel Sauer or a post-training lead.
  • Exact PhD completion status and date.

Why they are on the map

BFL co-founder (Stable Diffusion/SVD author); FLUX.1 Kontext author; senior research leader. Exact current title not verified. 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)