Rent a GPU

Can the Apple M2 Ultra (192GB) run Gemma 2 27B?

Apple M2 Ultra (192GB) × Gemma 2 27B8K context
YesRuns fully on the GPU
Memory at FP1658.1 GB of 144 GB
Weights 54.4 GBKV cache 3.1 GBRuntime 0.6 GB
QuantFP16Full quality
Memory58.1 GBof 144 GB usable
Speed~9tokens per second
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Yes. The Apple M2 Ultra (192GB) can run Gemma 2 27B fully in unified memory at FP16 (full quality), using about 58.1 GB of the 144 GB available, at roughly 9 tokens/second.

Try other context lengths

Run it with Ollama

With Ollama installed, one command downloads and starts Gemma 2 27B:

ollama run gemma2:27b

Ollama's default download is Q4_K_M (about 20.2 GB here). You have room for FP16, which is full quality: pick that GGUF in LM Studio or from Hugging Face.

Run it in the cloud

No Apple M2 Ultra (192GB)? Run Gemma 2 27B in the cloud.

Rent a cloud GPU by the hour. No upfront cost, stop any time.

Rent a GPU on RunPodRent a GPU on Vast.ai

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 144 GB the Apple M2 Ultra (192GB) can use. macOS lets the GPU use about 144 GB of the 192 GB unified memory by default.

QuantQualityMemoryFits?Speed
FP16Full quality58.1 GBFits~9 tok/s
Q8_0Near-lossless32.6 GBFits~17 tok/s
Q6_KExcellent26 GBFits~22 tok/s
Q5_K_MVery good23 GBFits~25 tok/s
Q4_K_MGood, most popular20.2 GBFits~29 tok/s
Q3_K_MNoticeably degraded17 GBFits~36 tok/s
Q2_KHeavily degraded13.9 GBFits~47 tok/s

GPUs that run Gemma 2 27B comfortably

Models that run well on the Apple M2 Ultra (192GB)

Frequently asked questions

Can Apple M2 Ultra (192GB) run Gemma 2 27B?

Yes. The Apple M2 Ultra (192GB) can run Gemma 2 27B fully in unified memory at FP16 (full quality), using about 58.1 GB of the 144 GB available, at roughly 9 tokens/second.

How much VRAM does Gemma 2 27B need?

About 20.2 GB at Q4_K_M with 8,192 tokens of context, and about 58.1 GB at full FP16 precision.

How fast is Gemma 2 27B on the Apple M2 Ultra (192GB)?

Roughly 9 tokens per second at FP16, estimated from the Apple M2 Ultra (192GB)'s 800 GB/s memory bandwidth.