FLUX.2 [dev]
A solid pick for self-hosting and full data control.
Overview
FLUX.2 [dev] is Black Forest Labs' open-weights image generation model, a 32-billion-parameter release from the German vendor designed to run on desktop-tier hardware. It's built for teams that want to prototype image generation on-premise, keeping prompts and outputs off third-party clouds. The catch: its flux-non-commercial licence blocks any production or commercial deployment without a separate agreement with Black Forest Labs. No benchmark figures accompany this release.
Skill profile
Not disclosed
Strengths
- Open-source and self-hostable
Limitations
- API pricing not disclosed
Who is it for
- you want to control cost or self-host
- you have GDPR constraints
Ideal use cases
- Local image prototyping
- On-prem creative R&D
- Confidential internal testing
- Design team experimentation
Access & availability
Key specifications
Privacy
Run it locally
| Quantization | Disk | RAM / VRAM | Typical hardware |
|---|---|---|---|
| Q4 · recommended | 18.2 GB | 22 GB | 32GB Mac / RTX 3090-4090 |
| Q8 · balanced | 34.2 GB | 40 GB | 64GB Mac / dual 24GB GPUs |
| FP16 · max quality | 64 GB | 72 GB | 96-128GB Mac Studio / 4× GPUs |
Estimates for a moderate context. Long contexts need more RAM (KV cache).
Deploy
Copy-ready commands generated from this card. Adjust context length and GPU count to your hardware.
OpenAI-compatible server for production on NVIDIA GPUs.
pip install vllm
vllm serve black-forest-labs/FLUX.2-dev \
--max-model-len 32768 \
--tensor-parallel-size 1 \
--dtype autoAdvanced data · for expertsArchitecture, modalities, detailed cost, full benchmarks▾
| Arena Elo | — |
| MMLU | — |
| GPQA | — |
| HumanEval | — |
| SWE-Bench | — |
| MATH | — |