Can an M3 Max 36GB Mac run Llama 3.3 70B?
M3 Max · 36 GB unified memory · 300 GB/s · MacBook Pro 14" (M3 Max), MacBook Pro 16" (M3 Max)
Llama 3.3 70B needs about 45.6 GiB but an M3 Max 36GB Mac has only ~30.6 GiB available for a model — you are 15.0 GiB short. macOS will swap to SSD and generation will crawl. There is a smaller quantisation that fits: see below.
Where the memory goes
Three things occupy memory when a model is loaded: the weights themselves, the KV cache that holds the conversation, and the runtime's own working buffers. Only the first is fixed — the KV cache grows with every token in your context.
| Quantisation | Weights | Total | Tok/s | Fits? |
|---|---|---|---|---|
| FP16 | 131.5 GiB | 141.4 GiB | 2 | No |
| Q8_0 | 69.9 GiB | 76.7 GiB | 3 | No |
| Q6_K | 54.2 GiB | 60.3 GiB | 4 | No |
| Q5_K_M | 46.8 GiB | 52.5 GiB | 5 | No |
| Q4_K_M | 40.3 GiB | 45.6 GiB | 6 | No |
| Q3_K_M | 32.1 GiB | 37.0 GiB | 7 | No |
| Q2_K | 24.7 GiB | 29.2 GiB | 9 | Yes, but tight |
The default. Best quality-per-gigabyte for most people.
How long a context fits
The KV cache is often what breaks a setup that looked fine at load time. Quantising it to 8-bit roughly halves its footprint with little measurable quality cost.
| Context | KV cache | Total | FP16 KV | Q8 KV |
|---|---|---|---|---|
| 2K | 0.6 GiB | 43.5 GiB | No | No |
| 4K | 1.3 GiB | 44.2 GiB | No | No |
| 8K | 2.5 GiB | 45.6 GiB | No | No |
| 16K | 5.0 GiB | 48.4 GiB | No | No |
| 32K | 10.0 GiB | 54.0 GiB | No | No |
| 128K | 40.0 GiB | 87.6 GiB | No | No |
What to run instead on an M3 Max 36GB Mac
These are the largest models that fit comfortably on this machine at Q4_K_M with an 8K context.
| Model | Size | Total | Tok/s | |
|---|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-32B | 32.8B | 22.4 GiB | 12 | Yes, comfortably |
| Qwen2.5 32B | 32.8B | 22.4 GiB | 12 | Yes, comfortably |
| Qwen2.5-Coder 32B | 32.8B | 22.4 GiB | 12 | Yes, comfortably |
| Qwen3 32B | 32.8B | 22.4 GiB | 12 | Yes, comfortably |
| Qwen3 30B-A3B | 30.5B | 19.8 GiB | 74 | Yes, comfortably |
| Gemma 3 27B | 27.4B | 21.1 GiB | 14 | Yes, comfortably |
Macs that run Llama 3.3 70B comfortably
The smallest configuration that handles it is the M1 Max 64GB (MacBook Pro 16" (M1 Max)).
| Mac | Memory | Bandwidth | Tok/s |
|---|---|---|---|
| M1 Max 64GB | 64 GB | 400 GB/s | 7 |
| M2 Ultra 64GB | 64 GB | 800 GB/s | 15 |
| M4 Pro 64GB | 64 GB | 273 GB/s | 5 |
| M4 Max 64GB | 64 GB | 546 GB/s | 10 |
| M2 Max 96GB | 96 GB | 400 GB/s | 7 |
| M3 Ultra 96GB | 96 GB | 819 GB/s | 15 |
| M3 Max 128GB | 128 GB | 400 GB/s | 7 |
| M4 Max 128GB | 128 GB | 546 GB/s | 10 |
| M2 Ultra 192GB | 192 GB | 800 GB/s | 15 |
| M3 Ultra 256GB | 256 GB | 819 GB/s | 15 |
| M3 Ultra 512GB | 512 GB | 819 GB/s | 15 |
Common questions
- Can an M3 Max 36GB Mac run Llama 3.3 70B?
- No. At Q4_K_M it needs about 45.6 GiB, and an M3 Max 36GB Mac has roughly 30.6 GiB available for a model after macOS takes its share.
- How fast is Llama 3.3 70B on an Mac?
- Around 6 tokens per second. Token generation on Apple Silicon is limited by memory bandwidth, and this machine has 300 GB/s.
- How much memory does Llama 3.3 70B need?
- 40.3 GiB for the weights at Q4_K_M, plus 2.5 GiB for an 8K-token KV cache and about 2.8 GiB of runtime overhead — 45.6 GiB in total.