p0pc0rn pack: Pictures (FLUX.2 [klein] 4B, 4-bit)

Converted and quantized by the p0pc0rn project from Black Forest Labs' FLUX.2 [klein] 4B. Modified by the p0pc0rn project: converted and quantized from the original weights.

Part Stored as Source
FLUX.2 [klein] 4B transformer 4-bit, groups of 32, asymmetric (globals 8-bit) black-forest-labs/FLUX.2-klein-4b-fp8, flux-2-klein-4b-fp8.safetensors (sha256 97ed34fe…)
Qwen3-4B text encoder (the 27 layers Klein reads) 4-bit, groups of 32 qwen_3_4b.safetensors (sha256 6c671498…), a bf16 copy of Qwen/Qwen3-4B
FLUX.2 autoencoder 16-bit floats flux2-vae.safetensors (sha256 d64f3a68…), the FLUX.2 VAE
Tokenizer vocabulary and merges verbatim Qwen3

Size: 4.62 GB in 59 shards. Licence: Apache-2.0 (LICENSE); notices in NOTICE.md. Black Forest Labs has not endorsed this pack.

About p0pc0rn

p0pc0rn is an on-device app by Joshua Hughes: pictures, short videos, songs and music videos, made by WebGPU on the phone. Where a licence asks for it (LTX-2, FLUX Non-Commercial), the app shows the licence and asks you to accept it before the first byte downloads; every licence is listed in the app under Settings › About › Licences and notices. The app checks every 16 MiB chunk against manifest.json (sha256) as it lands.

Format. pocketflux.pack/1: manifest.json lists every shard, every tensor (offset, shape, encoding) and the sha256 of every chunk; the .pfw shards are raw tensor data (group-wise 4-bit / 8-bit or 16-bit floats). Only p0pc0rn's WebGPU engine reads them. This is not a Transformers or Diffusers checkpoint.

Contact: X @p0pc0rnapp (official) or @joshhughes___ (Josh).

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for JoshHughes/p0pc0rn-klein-4b-q4

Finetuned
(96)
this model