DiffusionGemma 26B A4B
google/diffusiongemma-26B-A4B-it
Released2026-06-10Size25B · 3.8B activeArchitecturesliding-window attnLicenseapache-2.0AA rank#62 of 75
Files and memory at 8K context
| Quant | File | Files | + KV + runtime | Smallest fit | Source |
|---|---|---|---|---|---|
| Q4_K_M unsloth/diffusiongemma-26B-A4B-it-GGUF | 15.65 GiB | 1 | unknown | 20 GB GPU (cache unknown) | HF files |
| Q5_K_M unsloth/diffusiongemma-26B-A4B-it-GGUF | 17.83 GiB | 1 | unknown | 20 GB GPU (cache unknown) | HF files |
| Q8_0 unsloth/diffusiongemma-26B-A4B-it-GGUF | 25.03 GiB | 1 | unknown | 32 GB GPU (cache unknown) | HF files |
| MLX 4-bit mixed precision mlx-community/diffusiongemma-26B-A4B-it-4bit | 15.41 GiB | 4 | unknown | 24 GB Mac (cache unknown) | HF files |
| MLX 8-bit mixed precision mlx-community/diffusiongemma-26B-A4B-it-8bit | 26.05 GiB | 6 | unknown | 48 GB Mac (cache unknown) | HF files |
No verified file: FP16. Sizes are exact file bytes (text weights; projectors listed separately). “+” = recurrent state or other terms unknown. Smallest fit adds 0.5 GiB spare; Macs use 75% of memory.
Quality
Artificial Analysis Intelligence Index
9.5 AA estimate
Variant: DiffusionGemma 26B A4B · observed 2026-10-17 · source
Rank #62 of 75 researched 2026 models.
Arena text rating
Unknown
Scores describe the evaluated hosted variant, not each quantized file. AA and Arena are never averaged.
Reported speeds 0
No reported measurement yet. Submit one.
Community threads
- ADD FAST-dLLM mode for more speed Hugging Face · 2026-06-12
- RuntimeError: Tensor.item() cannot be called on meta tensors Hugging Face · 2026-06-11
Qualitative signals only; titles are not verified claims.