---
title: "Best Local LLMs for MacBook Pro M5 Pro 48GB (2026)"
canonical: https://llmcheck.net/best-llm/macbook-pro-m5-pro-48gb/
markdown: https://llmcheck.net/md/best-llm/macbook-pro-m5-pro-48gb.md
page_type: best-llm-mac
publisher: "LLMCheck (llmcheck.net)"
data_version: 2026-11-18
catalog_verified_through: 2026-11-18
updated: 2026-11-19
license: "CC BY 4.0"
license_url: https://creativecommons.org/licenses/by/4.0/
cite: "LLMCheck. (2026). Best Local LLMs for the MacBook Pro M5 Pro (48 GB). llmcheck.net. https://llmcheck.net/best-llm/macbook-pro-m5-pro-48gb/ (updated 2026-11-19). CC BY 4.0."
provenance:
  sourced: "Published figure — vendor or press review"
  community: "Linked individual run — blog, forum or submission"
  estimated: "Derived from published bandwidth model (https://llmcheck.net/methodology#estimation)"
---

# Best Local LLMs for the MacBook Pro M5 Pro (48 GB)

Markdown twin of https://llmcheck.net/best-llm/macbook-pro-m5-pro-48gb/, generated from the LLMCheck dataset (data version 2026-11-18) with the same figures as the page. Every speed names its provenance: **sourced** (a published vendor or press figure), **community** (one person's linked run) or **estimated** (the bandwidth model). LLMCheck is an index and runs no tests of its own. Licence CC BY 4.0: credit "LLMCheck (llmcheck.net)" and link the page.

According to the LLMCheck index (18 November 2026), the best local LLM for a MacBook Pro M5 Pro (48 GB) is **LFM2.5-24B-A2B** at 114 tok/s community-measured on an M5 Pro: the best-ranked model that reaches 15 tok/s on this Mac and has a standard Mac runtime. 71 of the 130 models in the index fit in 48 GB; 58 of them reach 15 tok/s here.

## This Mac

| Spec | Value |
| --- | --- |
| Unified memory | 48 GB |
| Memory bandwidth | 307 GB/s |
| Models that fit | 71 of 130 |
| Top speed | 387 tok/s (sourced) — Maple Preview 20B-A1B |
| Status | sold new by Apple |
| Price | $3,099 new (Apple list price, checked 2026-11-14) |

## Top 3 picks for the MacBook Pro M5 Pro (48 GB)

- **Best overall: [LFM2.5-24B-A2B](https://llmcheck.net/models/lfm25-24b-a2b-on-m5-pro/)** — 114 tok/s (community); 24B MoE, LFM Open, capability 35/50; agent-ready at 64K
- **Fastest: [Maple Preview 20B-A1B](https://llmcheck.net/models/maple-preview-20b-a1b-on-m5-pro/)** — 387 tok/s (sourced); 20B MoE, MIT, capability 20/50; agent-ready at 64K
- **Most capable: [Qwen3.8-27B](https://llmcheck.net/models/qwen38-27b-on-m5-pro/)** — 17 tok/s (sourced); 27.8B, Apache 2.0, capability 46/50; agent-ready at 64K

## Every model ranked for a MacBook Pro M5 Pro (48 GB) (top 12 of 71)

Ranked by LLMCheck suitability: capability plus min(15, tok/s × 0.12) on the M5 Pro (307 GB/s). Speeds are per-config figures for this Mac; a measured figure is the chip's run.

| # | Model | Size | License | tok/s | Provenance | Quant | 64K context | Capability | Note |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | [LFM2.5-24B-A2B](https://llmcheck.net/models/lfm25-24b-a2b-on-m5-pro/) | 24B MoE | LFM Open | 114 | community | oQ4e | agent-ready (~19 GB of 42 GB usable) | 35/50 | — |
| 2 | [Qwen3.8-27B](https://llmcheck.net/models/qwen38-27b-on-m5-pro/) | 27.8B | Apache 2.0 | 17 | sourced | 4-bit | agent-ready (~24 GB of 42 GB usable) | 46/50 | — |
| 3 | [Recursive Flint](https://llmcheck.net/models/recursive-flint-on-m5-pro/) | 39.8B MoE | Apache 2.0 | 92 | community | Q4_1 | agent-ready (~22 GB of 42 GB usable) | 37/50 | — |
| 4 | KAT-Coder-V2.5 | 35B MoE | Apache 2.0 | 72 | estimated | Q4_K_M | agent-ready (~26 GB of 42 GB usable) | 39/50 | — |
| 5 | [Qwen 3.6-35B-A3B](https://llmcheck.net/models/qwen-36-35b-a3b-on-m5-pro/) | 35B MoE | Apache 2.0 | 79 | community | Q4_K_M | agent-ready (~26 GB of 42 GB usable) | 38/50 | — |
| 6 | Nanbeige4.3-33B | 33.1B | Apache 2.0 | 12 | estimated | Q4_K_M | agent-ready (~31 GB of 42 GB usable) | 46/50 | slow (under 15 tok/s) |
| 7 | [Qwen 3.6-27B](https://llmcheck.net/models/qwen-36-27b-on-m5-pro/) | 27B | Apache 2.0 | 15 | estimated | Q4_K_M | agent-ready (~23 GB of 42 GB usable) | 44/50 | — |
| 8 | [Nemotron 3.5 Lightning](https://llmcheck.net/models/nemotron-35-lightning-on-m5-pro/) | 30B MoE | OpenMDW | 72 | community | Q4_K_M | agent-ready (~19 GB of 42 GB usable) | 36/50 | — |
| 9 | [North Mini Code 1.0](https://llmcheck.net/models/north-mini-code-10-on-m5-pro/) | 30.5B MoE | Apache 2.0 | 70 | community | 4-bit | agent-ready (~23 GB of 42 GB usable) | 36/50 | — |
| 10 | Gemma 4 26B-A4B | 26B MoE | Apache 2.0 | 78 | estimated | Q4_K_M | agent-ready (~20 GB of 42 GB usable) | 35/50 | — |
| 11 | [Raon-1 mini preview](https://llmcheck.net/models/raon-1-mini-preview-on-m5-pro/) | 28.8B MoE | Apache 2.0 | 86 | community | NVFP4 | agent-ready (~20 GB of 42 GB usable) | 34/50 | — |
| 12 | GLM-4.7-Flash | 31B MoE | MIT | 73 | estimated | Q4_K_M | agent-ready (~22 GB of 42 GB usable) | 35/50 | — |

Provenance: **sourced** = Published figure — vendor or press review; **community** = Linked individual run — blog, forum or submission; **estimated** = Derived from published bandwidth model (https://llmcheck.net/methodology#estimation).

## Context

A model is agent-ready at 64K when its weights plus a 65,536-token 16-bit KV cache fit in the 42 GB this Mac's GPU can use with the limit raised (57 of the 66 models here with a KV figure); an estimate from each model's config.json, not a measurement. A q8 KV cache halves the cache part. **GPU memory limit:** macOS lets the GPU use about 37.4 GB of this Mac’s 48 GB by default (macOS 26 and later, the only macOS it runs). Of the 71 models that fit, 7 need more than that at a short prompt: 3 need a raised limit and are tagged raise GPU limit (Nemotron 3.5 Bolt 74B-A9B and Llama 3.3 70B only past the safe setting below, as measured runs on this much memory show); 4 need more than even the raised limit and are tagged over GPU limit. `sudo sysctl iogpu.wired_limit_mb=43008` raises it to 42 GB, leaving 6 GB for macOS, until the Mac restarts (`sudo sysctl iogpu.wired_limit_mb=0` restores the default). [How the limit is read →](https://llmcheck.net/methodology#gpu-limit)

Estimated figures come from the bandwidth formula; against 163 measured runs its typical error is ±9% for dense models and ±15% for small MoEs (≤40B total). [How the error is computed →](https://llmcheck.net/methodology#typical-error)

## Quick start: LFM2.5-24B-A2B on the MacBook Pro M5 Pro

Ollama tag, verified on ollama.com 2026-11-14: `ollama run lfm2.5:24b`

Model card: https://huggingface.co/LiquidAI/LFM2.5-24B-A2B · Figures and setup: https://llmcheck.net/models/lfm25-24b-a2b-on-m5-pro/

## Related

- MacBook Pro M5 Pro (24 GB): 63 models fit — https://llmcheck.net/md/best-llm/macbook-pro-m5-pro-24gb.md
- MacBook Pro M5 Pro (64 GB): 77 models fit — https://llmcheck.net/md/best-llm/macbook-pro-m5-pro-64gb.md
- Best local LLM for every 32 GB Mac — https://llmcheck.net/md/best-llm/by-ram.md
- Every Mac — https://llmcheck.net/md/best-llm.md
- Full leaderboard — https://llmcheck.net/md/leaderboard.md


## Cite this page

LLMCheck. (2026). Best Local LLMs for the MacBook Pro M5 Pro (48 GB). llmcheck.net. https://llmcheck.net/best-llm/macbook-pro-m5-pro-48gb/ (updated 2026-11-19). CC BY 4.0.

How to cite: https://llmcheck.net/cite · Open data (CSV, JSON): https://llmcheck.net/data/ · Every Markdown mirror: https://llmcheck.net/md/index.md
