Ranking updated on 09/10/2026
The MacBook Pro M2 Pro / Max (16–96 GB, 200–400 GB/s) remains highly capable for local AI. 30B in Q4 is comfortable; 70B is accessible on Max 64+ GB.
Compare prices for MacBook Pro M5 Pro — 24 GB / 1 TB from our partner retailers (verified product pages):
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
Dense multimodal 27B released April 22, 2026. 262k ctx (1M YaRN). SWE-bench Verified 77.2%.
ollama run qwen3.6:27b
Qwen 3.8 27B: dense multimodal (text + vision), 262k context, ~16 GB Q4 VRAM (18 GB of Ollama weights). Apache 2.0, agentic coding and vision.
ollama run qwen3.8:27b
| Rank | Model | Params | Q4 VRAM | Context | License | On Apple M3 Pro (36 GB) |
|---|---|---|---|---|---|---|
| #1 | Gemma 4 26B-A4B MoE | 26B | 16 GB | 128 000 | Apache 2.0 | 8 tok/s · Q5_K_M |
| #2 | LLaDA 2.0 Uni 16B | 16B | 18 GB | 8 192 | Apache 2.0 | 25 tok/s · Q5_K_M |
| #3 | GLM 4.7 Flash | 31B | 19 GB | 128 000 | MIT | 15 tok/s · Q5_K_M |
| #4 | Granite 4.0 H-Small 32B-A9B | 32B | 19 GB | 128 000 | Apache 2.0 | 10 tok/s · Q5_K_M |
| #5 | Qwen 3 30B-A3B | 30B | 19 GB | 131 072 | Apache 2.0 | 15 tok/s · Q5_K_M |
| #6 | Qwen 3.6 27B | 27B | 16 GB | 262 144 | Apache 2.0 | 3 tok/s · Q5_K_M |
| #7 | Qwen 3.8 27B | 27B | 16 GB | 262 144 | Apache 2.0 | 9 tok/s · Q5_K_M |
Local AI on your Mac, fully explored: unified memory, MLX vs. GGUF, the right model for your chip, Ollama and LM Studio tuned for Apple Silicon.
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: 3-80B, with Q4_K_M fitting under 55 GB. Bonus 13-32B (M2 Max peak). Well-rated MoE models.
Criteria considered:
The scoring is fully transparent: see our methodology for details on VRAM/tokens/sec calculations.
16 GB M2 Pro MacBook Pro: which model?
Mistral 7B Q4 (~4.5 GB) or Qwen 3 8B Q4 (~5 GB) — 25–32 tok/s. For 13B, move up to an M2 Pro with 32 GB. See the MBP M2 guide.
MBP M2 Max 96 GB: is Llama 70B feasible?
Yes — Llama 3.3 70B Q4_K_M (~40 GB) runs at 8–12 tok/s. Slower than the M3 Max (200 GB/s vs 400 GB/s for memory), but usable for long-form work.
M2 Max vs RTX 4090?
Among 7–32B models that fit in 24 GB VRAM, the 4090 is 2–3× faster. The M2 Max pulls ahead once you exceed 24 GB (70B). See RTX 4090.
M2 vs. M3 vs. M4 Pro/Max?
On Mistral Small 24B Q4: M2 Max ≈ 18 tok/s, M3 Max ≈ 24 tok/s, M4 Max ≈ 28 tok/s. M2 Max remains competitive if you don't want to upgrade.