# MacBook Pro M4 Max 40c · 128 GB

> A laptop with twice Strix Halo's bandwidth. MLX and Metal llama.cpp are both mature.

Canonical: https://thelocalbench.com/hardware/macbook-pro-m4-max-128g · Data as of 2026-10-12 · The Local Bench (thelocalbench.com), data as of 2026-10-12, CC BY 4.0

Flags: [verified] checked against primary source · [sourced] cited · [estimate] formula · [sample] placeholder

## Specifications

| Field | Value | Flag | Source / note |
|---|---|---|---|
| memory_gb | 128 | sourced | https://support.apple.com/en-us/121553 |
| bandwidth_gbs | 546 | sourced | https://support.apple.com/en-us/121553 |
| fp16_tflops | 34 | sample | Sample FP16 dense compute; pending vendor-sheet citation. |
| load_watts | 140 | sample | Sample wall power under load. |
| price_usd | 4699 | sample | Sample price. |

## Models that run (default quant, 8k)

| Model | Quant | Memory GB | Verdict | Decode tok/s |
|---|---|---|---|---|
| [Llama 3.1 8B Instruct](/models/llama-3.1-8b.md) | Q4_K_M | 6.6 | fit | 41.0 [estimate] |
| [Qwen3 8B](/models/qwen3-8b.md) | Q4_K_M | 6.8 | fit | 39.4 [estimate] |
| [Qwen3 14B](/models/qwen3-14b.md) | Q4_K_M | 10.9 | fit | 23.8 [estimate] |
| [gpt-oss-20b](/models/gpt-oss-20b.md) | MXFP4 | 12.9 | fit | 107.3 [estimate] |
| [Mistral Small 3.2 24B](/models/mistral-small-3.2-24b.md) | Q4_K_M | 16.2 | fit | 15.7 [estimate] |
| [Gemma 3 27B IT](/models/gemma-3-27b.md) | Q4_K_M | 18.2 | fit | 13.9 [estimate] |
| [Qwen3 30B-A3B](/models/qwen3-30b-a3b.md) | Q4_K_M | 19.9 | fit | 87.4 [estimate] |
| [Qwen3 32B](/models/qwen3-32b.md) | Q4_K_M | 22.5 | fit | 11.2 [estimate] |
| [Llama 3.3 70B Instruct](/models/llama-3.3-70b.md) | Q4_K_M | 45.8 | fit | 5.4 [estimate] |
| [gpt-oss-120b](/models/gpt-oss-120b.md) | MXFP4 | 64.3 | fit | 79.9 [estimate] |