Can Inkling 2 Small run on the Mac M3 Ultra 512GB?
Yes. At 8K context the best fit is Q3_K_M, using about 263 GB of the 384 GB usable on the Mac M3 Ultra 512GB. Estimated speed ceiling: ~90 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 212 GB Fits | 213 GB Fits | 217 GB Fits | 235 GB Fits | 112 |
| 3-bit (Q3_K_M) • | 262 GB Fits | 263 GB Fits | 268 GB Fits | 286 GB Fits | 90 |
| 4-bit (Q4_K_M) • | 345 GB Fits | 346 GB Tight | 350 GB Tight | 368 GB Tight | 69 |
| 5-bit (Q5_K_M) • | 425 GB No | 426 GB No | 430 GB No | 448 GB No | — |
| 6-bit (Q6_K) • | 488 GB No | 489 GB No | 493 GB No | 511 GB No | — |
| 8-bit (Q8_0) • | 589 GB No | 590 GB No | 595 GB No | 613 GB No | — |
| 16-bit (F16/BF16) | 1108 GB No | 1109 GB No | 1113 GB No | 1131 GB No | — |
• real GGUF file size from bartowski/Inkling-2-Small-GGUF. Speed is a bandwidth ceiling (819 GB/s), not a benchmark. Methodology.
What fits instead
- No listed GPU with a list price runs Inkling 2 Small fully in memory. The smallest listed hardware that does is the Mac M3 Ultra 512GB (512 GB).
- Largest popular model that fits the Mac M3 Ultra 512GB: Nex N2.6 (321B, MoE) at Q8_0, about 319 GB at 8K context.
"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
Inkling 2 Small on every GPU and Mac · Every model that runs on the Mac M3 Ultra 512GB · 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 (thinkingmachines/Inkling-2-Small), Apple specs. Model data last checked 2026-11-17; refreshed hourly.