Ranking updated on 09/10/2026
The Mac mini M4 / M4 Pro (16-64 GB, 120-273 GB/s) is the best local inference server for performance per dollar in 2026. It is ventilated, quiet, and runs 24/7 without overheating.
Compare prices for Mac mini M5 Pro (24 GB / 512 GB) from our partner retailers (verified product pages):
Why this choice? Our complete guide to the Mac mini M5 Pro (24 GB / 512 GB) →
Which PC should you choose for your budget? Our picks from €800 to €3,500 →
Affiliate links — QuelLLM may earn a commission on purchases at no extra cost to you, which does not influence the ranking (established independently). As an Amazon Associate, BestLLMfor earns from qualifying purchases.
MoE variant of Gemma 4. 26B/4B active. Full multimodal (text+image+audio).
ollama run gemma4:26b
First open Apache 2.0 dLLM: MoE 16B/1B + 6.2B diffusion decoder. Unified text+vision. Released April 22, 2026.
# HuggingFace : inclusionAI/LLaDA2.0-Uni (Flash Attn 2 + CUDA 12.4 requis)
GLM-4.7-Flash (MoE 31B, ~3B active): the best code/VRAM ratio in the 30B class. MIT, 128k ctx, very fast on 3090/4090.
ollama run glm-4.7-flash
Mamba-2 + MoE 32B/9B hybrid. ~70% less RAM in long contexts. Apache 2.0.
ollama run granite4:small-h
MoE 30B/3B active hybrid thinking. MMLU 81.4, AIME24 80.4. 100+ languages.
ollama run qwen3:30b-a3b
MoE 30B (3.3B active parameters) specialized in agentic coding. Very fast locally, native 256k ctx, the benchmark for 16–24 GB via Ollama.
ollama run qwen3-coder:30b
Vision MoE with 30B/3B active. Vision sweet spot Qwen 3. 256k ctx.
ollama run qwen3-vl:30b
Korean agentic MoE with 30B/3B active parameters. Covers KR/EN/JP/ZH/TH/VI. Apache 2.0. MLA attention.
ollama pull hf.co/kakaoai/Kanana-2-30B-GGUF
| Rank | Model | Params | Q4 VRAM | Context | License | On Apple M4 Pro (48 GB) |
|---|---|---|---|---|---|---|
| #1 | Gemma 4 26B-A4B MoE | 26B | 16 GB | 128 000 | Apache 2.0 | 22 tok/s · Q8 |
| #2 | LLaDA 2.0 Uni 16B | 16B | 18 GB | 8 192 | Apache 2.0 | 60 tok/s · Q8 |
| #3 | GLM 4.7 Flash | 31B | 19 GB | 128 000 | MIT | 40 tok/s · Q8 |
| #4 | Granite 4.0 H-Small 32B-A9B | 32B | 19 GB | 128 000 | Apache 2.0 | 30 tok/s · Q8 |
| #5 | Qwen 3 30B-A3B | 30B | 19 GB | 131 072 | Apache 2.0 | 40 tok/s · Q8 |
| #6 | Qwen3-Coder 30B-A3B | 30B | 19 GB | 262 144 | Apache 2.0 | 40 tok/s · Q8 |
| #7 | Qwen 3 VL 30B-A3B | 30B | 19 GB | 262 144 | Apache 2.0 | 40 tok/s · Q8 |
| #8 | Kanana 2 30B-A3B Thinking | 30B | 18 GB | 131 072 | Apache 2.0 | 40 tok/s · Q8 |
Here's the ranking for your Mac mini M4. The Mac kit teaches you how to get the most out of it: push the GPU memory limit (ch. 2), choose between MLX and GGUF (ch. 3), and turn your Mac mini into an AI server for the whole house (ch. 12).
Free memo
Get the memo VRAM → best coding model → Ollama command (one screen, copy and paste). Then switch to the Copilote Local kit for a setup that actually works.
The Local Copilot kit — the Ollama + Cline + Aider configs are ready to paste, with tuned Modelfiles, troubleshooting, and lifetime online access →No spam. Unsubscribe in 1 click. Your data stays with us (never resold).
Your card → the best coding model to run locally, and the exact Ollama command:
| Your VRAM | Typical GPUs / Macs | Recommended coding model | Command Ollama |
|---|---|---|---|
| 8 GB | RTX 4060 / 3060 · M1-M2 16 GB | Qwen 3.5 9B (Q4, 6.6 GB — 256k context) | ollama run qwen3.5:9b |
| 12 GB | RTX 3060 12 GB / 4070 / 5070 | Qwen 3.5 9B (Q8, 11 GB) or Gemma 4 12B (7.6 GB) | ollama run qwen3.5:9b-q8_0 |
| 16 GB | RTX 5070 Ti / 4080 / 5080 · RX 9070 XT · M4 24 GB | Devstral 24B (Q4, 14 GB) — coding-agent specialist | ollama run devstral:24b |
| 24 GB | RTX 3090 / 4090 · RX 7900 XTX · M4 Pro 48 GB | Qwen 3.8 27B (Q4, 18 GB) — the “close to Copilot” option | ollama run qwen3.8:27b |
| 32 GB | RTX 5090 | Qwen 3.6 35B-A3B (Q4, 23 GB) — fast MoE | ollama run qwen3.6:35b |
| 48 GB+ | Mac M4 Max 64 GB · M2 Ultra 128 GB | Qwen3-Coder 30B-A3B (Q8, 32 GB — 256k context) | ollama run qwen3-coder:30b-a3b-q8_0 |
-base : ollama run qwen2.5-coder:7b-base — it’s still the reference for this specific use case. ⚠️ Qwen 3.8: its reasoning is set very high by default and it “overthinks” simple requests — lower it to low (or turn it off) on first launch. ⚠️ License trap: Codestral 22B = Mistral Non-Production License → prohibited for coding at work. Qwen 3.5/3.8, Gemma 4, and Devstral are Apache 2.0. 💡 Running out of memory? Keep ~1.5 GB of VRAM free for context, or drop down one quantization level.🔌 To connect it to VS Code: Cline (multi-file agent), Aider (CLI) or Tabby/Twinny (FIM autocomplete) — they all connect to Ollama locally. The kit Local Copilot — ready-to-paste configs + tested setup — is available: /copilote-local.
Filter: 1–70B whose Q4_K_M fits under 50 GB. Big bonus for 7–32B (peak M4 Pro) and MoE models (excellent on servers where first-token latency matters).
Criteria considered:
The scoring is fully transparent: see our methodology for details on VRAM/tokens/sec calculations.
Mac mini M4 16 GB: which model for a home LLM server?
Mistral 7B Q4 (~4.5 GB) or Qwen 3 8B Q4 (~5 GB) — 30-40 tok/s. Ideal for a Ollama server behind a €700 router. See the Mac mini M4 guide.
Mac mini M4 Pro 48 GB: can it handle 32B?
Yes — Qwen 3 32B Q4 (~17 GB) at 22-28 tok/s, Qwen 3 30B-A3B (MoE, ~17 GB) at 50-60 tok/s. This is the best Mac mini for local AI in 2026.
Mac mini M4 vs RTX 5070 Ti?
RTX 5070 Ti (16 GB GDDR7, ~750 GB/s) is ~2× faster on 7-13B models. The Mac mini pulls ahead once you exceed 16 GB (30B models). And with silence plus power consumption < 100W, it's hard to beat for 24/7 use.
What’s the ideal Mac mini M4 configuration?
M4 Pro 48 GB / 1 TB SSD = ~2300 € — sweet spot. M4 Pro 64 GB opens the door to Llama 70B Q3. The base M4 with 16 GB remains excellent for 7-8B models as an entry-level server.