UIGEN-T3-4B-Preview-MAX-GGUF

UIGEN-T3-4B-Preview-MAX is a model built on the Qwen3 architecture, designed to support both UI components and full web page generation. It features a hybrid reasoning system that allows users to switch between /think and /no_think modes for flexible generation. The model is powered by UIGenEval, the first benchmark specifically created for evaluating UI generation. Released for research and non-commercial use, commercial applications require contacting the team to participate in a pilot program.

Model Files

Filename	Size	Format	Description
UIGEN-T3-4B-Preview-MAX.BF16.gguf	8.05 GB	BF16	Brain Float 16-bit quantization
UIGEN-T3-4B-Preview-MAX.F16.gguf	8.05 GB	F16	Half precision (16-bit) floating point
UIGEN-T3-4B-Preview-MAX.F32.gguf	16.1 GB	F32	Full precision (32-bit) floating point
UIGEN-T3-4B-Preview-MAX.Q2_K.gguf	1.67 GB	Q2_K	2-bit quantization with K-quant
UIGEN-T3-4B-Preview-MAX.Q3_K_L.gguf	2.24 GB	Q3_K_L	3-bit quantization (Large) with K-quant
UIGEN-T3-4B-Preview-MAX.Q3_K_M.gguf	2.08 GB	Q3_K_M	3-bit quantization (Medium) with K-quant
UIGEN-T3-4B-Preview-MAX.Q3_K_S.gguf	1.89 GB	Q3_K_S	3-bit quantization (Small) with K-quant
UIGEN-T3-4B-Preview-MAX.Q4_K_M.gguf	2.5 GB	Q4_K_M	4-bit quantization (Medium) with K-quant
UIGEN-T3-4B-Preview-MAX.Q4_K_S.gguf	2.38 GB	Q4_K_S	4-bit quantization (Small) with K-quant
UIGEN-T3-4B-Preview-MAX.Q5_K_M.gguf	2.89 GB	Q5_K_M	5-bit quantization (Medium) with K-quant
UIGEN-T3-4B-Preview-MAX.Q5_K_S.gguf	2.82 GB	Q5_K_S	5-bit quantization (Small) with K-quant
UIGEN-T3-4B-Preview-MAX.Q6_K.gguf	3.31 GB	Q6_K	6-bit quantization with K-quant
UIGEN-T3-4B-Preview-MAX.Q8_0.gguf	4.28 GB	Q8_0	8-bit quantization

Quants Usage

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):