Model Comparison

ggml-org/gemma-4-31b-it-ggufvsunsloth/z-image-turbo-gguf

Side-by-side comparison of ggml-org/gemma-4-31b-it-gguf and unsloth/z-image-turbo-gguf: downloads, license, context length, tasks, and benchmarks.

ggml-org/gemma-4-31b-it-gguf

ggml-org · —

# gemma-4-31B-it-GGUF Recommended way to run this model: ``sh llama-server -hf ggml-org/gemma-4-31B-it-GGUF `` Then, access http://localhost:8080

# Use bfloat16 for optimal performance on supported GPUs pipe = ZImagePipeline.from_pretrained( "Tongyi-MAI/Z-Image-Turbo", torch_dtype=torch.bfloat16, low_cpu_mem_usage=False, ) pipe.to("cuda") # [Optional] Attention Backend # Diffusers uses SDPA by default. Switch to Flash Att…

Side-by-side Specifications

	ggml-org/gemma-4-31b-it-gguf	unsloth/z-image-turbo-gguf
Author	ggml-org	unsloth
Pipeline Task	—	text-to-image
Library	—	ggml
Downloads	52,521	44,268
Likes	34	133
License	Unknown	Unknown
Context Length	—	—
Created	2026-04-01	2025-12-21
Last Modified	2026-04-12	2026-01-08
Tags	ggufbase_model:google/gemma-4-31B-itbase_model:quantized:google/gemma-4-31B-itendpoints_compatibleregion:usconversational	ggmlggufquantizedunslothtext-to-imageenarxiv:2511.22699arxiv:2511.22677arxiv:2511.13649base_model:Tongyi-MAI/Z-Image-Turbo

View full details: ggml-org/gemma-4-31b-it-gguf · unsloth/z-image-turbo-gguf