RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

Can Granite 4.0 H Small run on the Mac M1 Max 64GB?

Yes. At 8K context the best fit is Q8_0, using about 32.6 GB of the 48.0 GB usable on the Mac M1 Max 64GB. Estimated speed ceiling: ~43 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •11.6 GB Fits11.7 GB Fits12.1 GB Fits13.6 GB Fits124
3-bit (Q3_K_M) •15.0 GB Fits15.0 GB Fits15.4 GB Fits16.9 GB Fits95
4-bit (Q4_K_M) •18.8 GB Fits18.9 GB Fits19.2 GB Fits20.7 GB Fits75
5-bit (Q5_K_M) •21.9 GB Fits22.0 GB Fits22.4 GB Fits23.9 GB Fits64
6-bit (Q6_K) •25.3 GB Fits25.4 GB Fits25.7 GB Fits27.2 GB Fits55
8-bit (Q8_0) •32.6 GB Fits32.6 GB Fits33.0 GB Fits34.5 GB Fits43
16-bit (F16/BF16) •60.7 GB No60.7 GB No61.1 GB No62.6 GB No—

• real GGUF file size from ibm-granite/granite-4.0-h-small-GGUF. Speed is a bandwidth ceiling (400 GB/s), not a benchmark. Methodology.

What fits instead

"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.

More

Granite 4.0 H Small on every GPU and Mac · Every model that runs on the Mac M1 Max 64GB · Apple Silicon guide · Try other settings in the calculator

Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs

Data: Hugging Face Hub (ibm-granite/granite-4.0-h-small), Apple specs. Model data last checked 2026-11-18; refreshed hourly.