GoogleGemma 4 26B-A4B on NVIDIA GeForce RTX 3090
Yes — RTX 3090 runs Gemma 4 26B-A4B excellently at Q5_K_M — 213 tok/s. 24 GB VRAM with plenty of headroom.
Model Size
25.2B
Device VRAM
24 GB
Bandwidth
936 GB/s
Quantization
Q5_K_M
Performance by Quantization
OwnRig currently has one published compatibility entry for Gemma 4 26B-A4B on NVIDIA GeForce RTX 3090 at Q5_K_M. This is the best supported pairing we can stand behind today.
| Quantization | Speed | TTFT | Fits in VRAM | Rating | Confidence |
|---|---|---|---|---|---|
| Q5_K_M | 213 tok/s | 50ms | ✓ Yes | Excellent | estimated |
Notes
Q5_K_M
25.2B model. Q5_K_M 19.32GB on 24GB discrete_gpu.
About Gemma 4 26B-A4B
Gemma 4 26B-A4B (25.2B) is a chat, coding, reasoning, multi-purpose model. Mixture-of-Experts architecture: 25.2B total parameters but only 3.8B active per token (8 selected + 1 shared expert per layer, out of 128 total). Hybrid dense+sparse FFN design. Inference throughput closer to a 4B dense model; quality closer to a 27B dense model. 256K context window. Benchmarks: 88.3% AIME 2026, 82.6% MMLU Pro, 77.1% LiveCodeBench. All 25.2B weights must be loaded into VRAM despite sparse activation; fits on 24 GB GPUs at Q4_K_M. Apache 2.0 licensed.
View all Gemma 4 26B-A4B hardware options →About NVIDIA GeForce RTX 3090
NVIDIA GeForce RTX 3090 has 24 GB at 936 GB/s. Street price: $899.
See all models NVIDIA GeForce RTX 3090 can run →Builds with NVIDIA GeForce RTX 3090
Extreme AI Workstation
Dual GPUs that run the biggest AI models at a smart price
Runs 8 models
High-End Home AI Server
Your household's private AI: chatbots, code tools, and more
Runs 12 models
Mid-Range Home AI Server
Serve multiple AI models to every device at home
Runs 9 models
Estimate method: Estimated from MoE active params (3.8B), quantization, and device bandwidth with 0.65 efficiency factor. Reference hardware source: huggingface.co (2026-04-04)
Performance varies by driver version, inference engine, quantization method, context length, and system configuration. Figures shown are estimates based on community benchmarks and may not reflect your exact setup. Product names are trademarks of their respective owners. OwnRig is independent and not affiliated with any hardware or AI model provider.