Model Detail
diffusiongemma-26B-A4B-it
—diffusiongemma-26B-A4B-it is a multimodal model with 26B parameters released by Google. The model is registered under the image-text-to-text pipeline tag on Hugging Face, distributed under the permissive apache-2.0 license.
diffusiongemma-26B-A4B-it ships with 26B parameters. Total weight footprint is approximately 25.8 GB, which is the relevant figure when planning local-inference VRAM. The apache-2.0 license is permissive, allowing commercial deployment and derivative work without per-seat fees, though attribution requirements still apply.
Downloads of diffusiongemma-26B-A4B-it have moved +14.2% over the trailing seven days, -66.1% over the trailing thirty days. The trend is mildly positive, consistent with a model that is being picked up incrementally rather than going viral. These numbers are signal, not guarantee — week-over-week download counts on Hugging Face also reflect mirror traffic, CI scrapes, and one-off benchmarking runs.
diffusiongemma-26B-A4B-it is best fit for mixed text-and-image reasoning tasks such as document understanding. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.
How Transparent is DiffusionGemma?
arXiv:2606.20560v1 Announce Type: cross Abstract: LLM reasoning transparency is a critical affordance for understanding model decisions, mitigating misuse and misalignment, and debugging surprising model behaviors. However, DiffusionGemma performs a larger fraction of its computation in a continuous
Neither Parallel Nor Sequential: How DiffusionGemma Actually Commits Tokens
arXiv:2606.14620v1 Announce Type: new Abstract: Open diffusion language models are marketed as parallel, non-autoregressive decoders, yet the order in which a shipped checkpoint actually commits its tokens is almost never measured. We instrument DiffusionGemma 26B, a masked discrete-diffusion mixtur
