The Free & Uncensored Version of MidJourney! (FLUX.1)

The Free & Uncensored Version of MidJourney! (FLUX.1)

by Matt Wolfe

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Matt Wolfe 👥 1.0M 📅 August 6, 2024 ⏱ 18 min 👁 115K 📄 news review 🧭 2026-08-28
Available in: English (current) Français

Keywords

FLUX.1AI image generationopen sourceMidJourneyDALL-E 3

Summary

Matt Wolfe introduces FLUX.1, a new AI image generation model from Black Forest Labs, founded by former Stable Diffusion team members. The video explains the three variants: Schnell (open-source, Apache 2.0), Dev (non-commercial), and Pro (enterprise). Wolfe demonstrates how to use FLUX.1 for free via Hugging Face spaces and the Glif platform, which even offers access to the Pro model. He interviews Miguel (Angry Penguin) to discuss strengths and weaknesses. FLUX.1 excels at realism and text generation, but lags behind MidJourney in stylistic illustrations and behind DALL-E 3 in prompt adherence. The model is relatively uncensored, allowing generation of copyrighted characters and political figures, though it still has an NSFW filter. Wolfe compares FLUX.1 to MidJourney and DALL-E 3 with several test prompts, showing that while not yet superior, it is close and has great potential due to its open-source nature. He also mentions that FLUX.1 will later be extended to text-to-video. The video concludes that FLUX.1 could soon combine the best of MidJourney, DALL-E 3, and Stable Diffusion, and encourages viewers to explore it.

175 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides valuable hands-on information about FLUX.1, including practical usage tips and comparisons with other models. The argumentation is based on direct testing and expert input, which adds credibility. However, the tests are not systematic and the conclusions are somewhat subjective, relying on visual inspection rather than quantitative metrics. The host’s enthusiasm is evident, but he also acknowledges limitations, providing a balanced view.

Scientific Rigor, Source Quality, Title Accuracy

The video references the Hugging Face page of Black Forest Labs and the Glif platform, which are directly relevant. The host also mentions the team’s background, but does not provide specific citations for the claims about the model’s architecture. The title accurately reflects the content, and the video is well-structured. The presence of a promotional segment for FutureTools is noted but does not detract from the overall quality.

147 words

Title / Content Match

The title accurately reflects the content: the video presents FLUX.1 as a free and less censored alternative to MidJourney, and the content matches this claim.

Quality & Reliability

7/10

The video provides a hands-on review of FLUX.1, with practical tests and comparisons to MidJourney and DALL-E 3. The host interviews an expert (Angry Penguin) and provides clear information about the models' capabilities and limitations. However, the evaluation is largely anecdotal and lacks systematic benchmarking, and the promotional tone and lack of detailed technical analysis slightly reduce the reliability score.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

  • Black Forest Labs on Hugging Face — Official page for FLUX.1 models, where users can try the Schnell and Dev versions.
  • Glif — Platform used in the video to access FLUX.1 Pro for free and build workflows.
  • FutureTools — Matt Wolfe's website for exploring AI tools, mentioned in the video.
  • FutureTools Newsletter — Newsletter mentioned in the video for staying updated on AI tools.
  • FutureTools Discord — Discord community for FutureTools, mentioned in the video.

Concurring Sources

External References

Contribution & Novelties

The video provides an early, accessible overview of FLUX.1, highlighting its open-source nature and potential to disrupt the AI image generation landscape. It offers practical tips for using the model and compares it directly with MidJourney and DALL-E 3, giving viewers a clear sense of its current capabilities.

Pour aller plus loin :

  • Black Forest Labs — Official website of the company behind FLUX.1, providing more details on the models and their roadmap.
  • Apache 2.0 License — The open-source license under which FLUX.1 Schnell is released, explaining usage rights.
  • Stable Diffusion — Background on the predecessor models and the team’s history, useful for understanding FLUX.1’s lineage.

106 words

Radar Profile

The radar profile shows high scores in information quantity and quality, reflecting the video's comprehensive coverage. The technical level is moderate, suitable for a broad audience, while reliability is solid due to hands-on testing and expert input.

Reliability 7/10

💬 Très positif. Sur les 30 commentaires analysés, la grande majorité exprime un enthousiasme marqué pour FLUX.1, saluant sa qualité, son ouverture et son potentiel, avec quelques remarques constructives sur les comparaisons et les améliorations possibles.