← The openness index

image · Openness profile

HiDream-I1 (Full / Dev / Fast)

HiDream.ai · snapshot 2026-11-07

open image generative foundation model built on a sparse diffusion transformer

Tier 3 Practical open-weight

What the publisher claims: open-source, MIT licensed

The one-liner: MIT-licensed image weights with a published technical report behind them.

The six questions

A AccessScore 3 of 3

Weights for the Full, Dev, and Fast variants are downloadable from the publisher's Hugging Face repositories.

P PermissionScore 3 of 3

Both the repository code and the model weights are released under the MIT License, an OSI-approved permissive license.

D DeploymentScore 3 of 3

Runs locally on consumer GPUs with published inference code and Diffusers-compatible checkpoints.

T TransparencyScore 2 of 3

A technical report documents the sparse diffusion transformer architecture and training approach; the training corpus itself is not released.

R ReproducibilityScore 1 of 3

Limited: strong disclosure of method, but no reproducible data or full recipe.

G Governance & ParticipationScore 2 of 3

Academic-team repo with merged community PRs; no formal governance structure.

License

MIT License (code and model weights) OSI-approved — license source ↗

What is actually released

Why it matters

A useful counterweight to the FLUX and Stability families: the same distributable-weight practice with a genuinely permissive license attached.

Sources

Derived from the openness dataset behind But Is It Open? — snapshot 2026-11-07 ↗. Scores are qualitative readings of the cited evidence, not verdicts; tier measures openness, not intelligence.

← The openness indexJoin the newsletter ↗