HF Model & Benchmark Matrix

dflash_gemma4_12b_block7

deepseek-ai/dflash_gemma4_12b_block7

kind

original

quant

bf16 (16-bit)

family

gemma

modality

—

root model

deepseek-ai/dflash_gemma4_12b_block7

provider

deepseek-ai

license

—

language

—

context

—

library

—

vRAM (GB)

—

size on disk

—

downloads

1,103

likes

7

created

—

last modified

2026-06-28T12:34:57+00:00 (2026)

huggingface

https://huggingface.co/deepseek-ai/dflash_gemma4_12b_block7

tags

safetensorsgemma4_textregion:us

Benchmark scores (0)

no scores extracted for this variant.

Discussions (0)

none scraped.