Models › gemma-4-26B-A4B-it-qat-q4_0-unquantized
gemma-4-26B-A4B-it-qat-q4_0-unquantized: VRAM needs and speed by GPU
At UD-Q4_K_XL, gemma-4-26B-A4B-it-qat-q4_0-unquantized needs about 15.7 GB of VRAM. No measured runs yet. Speeds below are estimates. Submit a run.
25.23B parameters (mixture-of-experts, ~3.9B active) · license: apache-2.0 · 262,144-token context · GGUF published Jun 5, 2026 · GGUF files on Hugging Face
Memory needed per quantization
| Quant | File size | Needs (estimate) | Smallest card it fits |
|---|---|---|---|
| UD-Q4_K_XL | 14.2 GB | 15.7 GB | 24 GB cards |
Expected speed on popular hardware
| Hardware | Best quant that fits | Expected tok/s |
|---|---|---|
| NVIDIA RTX 3060 | Does not fit | |
| NVIDIA RTX 4060 Ti | Does not fit | |
| NVIDIA RTX 3090 | UD-Q4_K_XL | 150 estimate |
| NVIDIA RTX 4090 | UD-Q4_K_XL | 166 estimate |
| NVIDIA RTX 5090 | UD-Q4_K_XL | 262 estimate |
| AMD Radeon RX 7900 XTX | UD-Q4_K_XL | 167 estimate |
| Apple M4 Pro (20-core GPU) (64 GB) | UD-Q4_K_XL | 52.8 estimate |
| Apple M3 Max (40-core GPU) (64 GB) | UD-Q4_K_XL | 77.1 estimate |
| AMD Ryzen AI Max+ 395 (Radeon 8060S) | UD-Q4_K_XL | 47.4 estimate |
Too big for your card?
Run gemma-4-26B-A4B-it-qat-q4_0-unquantized on a rented 24-80 GB GPU by the hour.
Rent a bigger GPU on RunPodAffiliate link: if you sign up, RigBench may earn a referral credit. It never affects rankings or numbers.