FLUX.2-klein-4B-pouw

A self-contained pouw model, based on black-forest-labs/FLUX.2-klein-4B. It bundles the full base weights (apache-2.0) together with the metadata that makes it mine MatMulToken Proof-of-Useful-Work while it serves — pull this one repo and it runs, no second download.

MatMulToken's mining is output-preserving: generation is bit-identical to the base model. The eligible transformer matmuls (in_features == common_dim = 3072) are reused as PoW lottery tickets — you serve real images and mine on the same compute, no second matmul.

It is GPU-agnostic (portable Triton/PyTorch kernels, no CUDA build): RTX 3090 (sm86) → 5090 → H100 → B200, same code.

Mining shape

field	value
base model	black-forest-labs/FLUX.2-klein-4B
modality	image
common_dim	3072
rank	32
mine_layers	16 (overhead dial; layer count)
pipeline	diffusers

Use

# install the MatMulToken miner into your serving venv (see the MatMulToken repo)
#   uv pip install --no-deps <matmul_mining wheel> -e miner-base -e vllm-matmul ...
from vllm_matmul import matmultoken_load
b = matmultoken_load("Matmultoken/FLUX.2-klein-4B-pouw", gateway=False)   # gateway=True for the live chain
b["pipe"]("a single matmul on a clean white desk, studio light")   # serves AND mines
print("wrapped", b["wrapped"], "mining linears; common_dim", b["common_dim"])

gateway=False attaches an idle local job (for testing the mining path); gateway=True connects to a running MatMulToken gateway for the live block template / target.

Notes

• The live PoW job + difficulty target always come from the chain at runtime — never baked into this repo. GPU kernels compile per-arch on first run (one-time, cached on disk).
• Published under the Matmultoken organization. The base weights (apache-2.0) are bundled in this repo at a pinned snapshot for a reproducible mining shape; the original model's LICENSE and attribution are preserved in-repo.

Generated by MatMulToken publish_pouw_models.py. License: MIT.