Yes — Grok 4 Open (100B) runs at 40 tok/s on M4 Ultra with 192 GB RAM using Q4_K_M quantization via MLX. First token latency is 0.8s. A capable open-source LLM with 100B parameters.
Want to run Grok 4 Open faster — or step up to bigger models? Find your Mac →The LLMCheck index estimates Grok 4 Open on M4 Ultra using our published methodology: Q4_K_M quantization, memory-bandwidth scaling, and cross-referenced third-party benchmarks where available. Figures are transparent estimates — own this config? Submit a real benchmark →
| Metric | Value |
|---|---|
| Tokens per second | 40 tok/s |
| Time to first token | 0.8s |
| Quantization | Q4_K_M |
| Minimum RAM | 192 GB |
| Recommended engine | MLX |
| Parameters | 100B |
| Benchmark date | 2026-06 |
Q4_K_M 100B MLX M4 Ultra
The recommended engine for Grok 4 Open on M4 Ultra is MLX. Install with pip and pull the model:
Alternatively, you can use Ollama for a simpler setup:
| Chip | Speed | First Token | Min RAM | Engine |
|---|---|---|---|---|
| M5 Max | 32 tok/s | 1.0s | 128 GB | MLX |
| M5 Max | 28 tok/s | 1.2s | 64 GB | Ollama |
To run Grok 4 Open on M4 Ultra you need:
Grok 4 Open needs about 192 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep our benchmarks free.
See how Grok 4 Open stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard