text model · Gemma · macOS
Can I run Gemma 4 26B-A4B on Apple M5 (32GB)?
Yes. Gemma 4 26B-A4B runs on Apple M5 (32GB) at Q4_K_M (~19 GB of ~21 GB usable).
Fits at Q4_K_M (~19 GB of ~21 GB usable) but with little headroom. Close other apps, or drop to a 2k context to free about a gigabyte.
That figure is at a 4k context and moves about ±15% as context length changes. Apple M5 (32GB) leaves ~2 GB of headroom.
- Q4_K_M needed
- ~19 GB
- Usable on device
- ~21 GB
- Device memory
- 32 GB
- Best quant
- Q4_K_M
Which quant fits
Run it
Pick your tool. All 3 load the same Q4_K_M weights.
ollama run gemma4:26b-a4b llama-cli -hf bartowski/google_gemma-4-26B-A4B-it-GGUF:Q4_K_M lms get bartowski/google_gemma-4-26B-A4B-it-GGUF - Parameters
- 26.5B (MoE, 3.8B active)
- Q4_K_M size
- 17.04 GB
- Q8_0 size
- 26.86 GB
- Context
- 256k
- Ollama tag
- gemma4:26b-a4b
- Memory
- 32 GB unified
- Usable for weights
- ~21 GB
- Best runtime
- MLX direct / Ollama (MLX backend)
You could also run
Run Gemma 4 26B-A4B on other hardware
FAQ
Can Apple M5 (32GB) run Gemma 4 26B-A4B?
Yes. Gemma 4 26B-A4B runs on Apple M5 (32GB) at Q4_K_M (~19 GB of ~21 GB usable).
How much memory does Gemma 4 26B-A4B need?
It is a tight fit on Apple M5 (32GB). At Q4_K_M the weights are ~17.04 GB; with KV cache and runtime overhead, budget ~19 GB at a 4k context. It is a Mixture-of-Experts model (26.5B total / 3.8B active), so all experts must stay in memory; memory tracks total params, not active params.
What is the best tool to run Gemma 4 26B-A4B on macOS?
LM Studio for a simple setup; mlx-lm for the most speed. vLLM is NOT a Mac tool, it is a CUDA/Linux serving engine. Unified memory is not a fixed VRAM slice; ~70% is usable for weights.
Embed this
A live badge for your model card or README, updated as the data is. [](https://localmodel.run/can-i-run/gemma-4-26b-a4b/apple-m5-32gb) Sources
Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.