DiffusionGemma 26B-A4B
DiffusionGemma 26B-A4B is an open-weights language model from DeepMind, released June 10, 2026. It has 26B parameters (mixture-of-experts, 4B active). The context window is 256K tokens.
Check it against your own PC →
Can you run DiffusionGemma 26B-A4B locally?
The smallest common GPU it runs on is 6 GB of VRAM (with 32 GB of RAM): Q6_K, GPU + experts in RAM, roughly 14 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Runs · GPU + CPU experts | Q6_K (≈ lossless) | 5.2 GB | 17.5 GB | ~14 |
| 8 GB VRAM | Runs · GPU + CPU experts | Q6_K (≈ lossless) | 7.2 GB | 15.5 GB | ~16 |
| 12 GB VRAM | Runs · GPU + CPU experts | Q6_K (≈ lossless) | 11.2 GB | 11.5 GB | ~21 |
| 16 GB VRAM | Runs on GPU | IQ4_XS (good) | 15.2 GB | – | ~120 |
| 24 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 22.7 GB | – | ~150 |
| 32 GB VRAM | Runs on GPU | Q6_K (≈ lossless) | 22.7 GB | – | ~250 |
Ready-made quantizations on Hugging Face: unsloth/diffusiongemma-26B-A4B-it-GGUF · AlexAtomic/diffusiongemma-26B-A4B-it-GGUF · aogavrilov/diffusiongemma-agent-iq3-cuda13 · DevQuasar/google.diffusiongemma-26B-A4B-it-GGUF · FreedomAISVR/DiffusionGemma-26B-A4B-it-NVFP4-GGUF · cyankiwi/diffusiongemma-26B-A4B-it-AWQ-INT4
Benchmark results
Epoch AI has not published results for DiffusionGemma 26B-A4B yet. New models are usually evaluated within days to weeks.
Links
Announcement · Hugging Face: google/diffusiongemma-26B-A4B-it