What can Apple Mac Studio (M3 Ultra) run for coding?
Build: Apple Mac Studio (M3 Ultra) + — + 192 GB RAM (macos)
Runs comfortably130 models
Ranked by fit for coding use case + predicted speed. Click a row for VRAM breakdown.
Quant: Q4_K_MContext: 8,192VRAM: 8.4 GBHeadroom: 175.6 GBollama run codegemma:7b124tok/sEstimated
ollama run codegemma:7bQuant: Q4_K_MContext: 8,192VRAM: 18.5 GBHeadroom: 165.5 GBollama run deepseek-coder-v2:16b54tok/sEstimated
ollama run deepseek-coder-v2:16bQuant: Q8_0Context: 8,192VRAM: 9.5 GBHeadroom: 174.5 GBollama run qwen2.5:7b70tok/sEstimated
ollama run qwen2.5:7bQuant: Q8_0Context: 8,192VRAM: 35.7 GBHeadroom: 148.3 GBollama run codestral:22b22tok/sEstimated
ollama run codestral:22bQuant: Q8_0Context: 8,192VRAM: 13.1 GBHeadroom: 170.9 GBollama run qwen3:8b61tok/sEstimated
ollama run qwen3:8bQuant: Q8_0Context: 8,192VRAM: 23.3 GBHeadroom: 160.7 GBollama run qwen3:14b35tok/sEstimated
ollama run qwen3:14bQuant: Q8_0Context: 8,192VRAM: 24.0 GBHeadroom: 160.0 GBollama run qwen2.5:14b35tok/sEstimated
ollama run qwen2.5:14bQuant: Q8_0Context: 8,192VRAM: 38.4 GBHeadroom: 145.6 GBollama run qwen2.5-coder:32b15tok/sEstimated
ollama run qwen2.5-coder:32bQuant: Q8_0Context: 8,192VRAM: 49.1 GBHeadroom: 134.9 GBollama run qwen3:30b16tok/sEstimated
ollama run qwen3:30bQuant: Q8_0Context: 8,192VRAM: 50.7 GBHeadroom: 133.3 GBollama run gemma4:31b16tok/sEstimated
ollama run gemma4:31bQuant: Q8_0Context: 8,192VRAM: 52.2 GBHeadroom: 131.8 GBollama run qwen3:32b15tok/sEstimated
ollama run qwen3:32bQuant: Q8_0Context: 8,192VRAM: 52.2 GBHeadroom: 131.8 GBollama run qwen2.5:32b15tok/sEstimated
ollama run qwen2.5:32bRuns with tradeoffs1 models
Tight VRAM, partial CPU offload, or context-limited.
Quant: Q4_K_MContext: 2,048VRAM: 183.3 GBHeadroom: 0.7 GB- • Tight VRAM fit — only 0.7 GB headroom left for context growth
3tok/sEstimated
- • Tight VRAM fit — only 0.7 GB headroom left for context growth
What if you upgraded?
Hypothetical scenarios. We re-ran the compatibility engine for each.
Move up an Apple memory tier
Varies by configuration. Check Apple's current pricing
On Apple Silicon, more unified memory is the only path forward — VRAM and system RAM are the same pool.
Won't runtop 5 popular models
Need more memory than you have. Shown for orientation.
Needs ~1024 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
—
Needs ~1024 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
Needs ~420 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
—
Needs ~420 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
Needs ~192 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
—
Needs ~192 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
Needs ~420 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
—
Needs ~420 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
Needs ~700 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
—
Needs ~700 GB unified memory minimum at smallest quant; you have 184 GB available after OS overhead.
Not enough VRAM for these models? Rent a GPU by the hour
The models above need more memory than 192 GB unified memory. You can rent a GPU with enough VRAM by the hour to run one of them, or to test it before you buy hardware. For daily use, owning a card usually costs less over time.
RunPod
billed per GPU-hourHourly GPU pods, in community-cloud and secure-cloud tiers.
The two tiers run on different hosts and terms. Check which tier a pod is on before a long job.
Rent on RunPodVast.ai
billed per GPU-hourMarketplace of host-listed GPUs; price and reliability vary by host.
Host quality varies. Filter for verified hosts and check the host's reliability before a long run.
Rent on Vast.aiReferral links: RunLocalAI may earn a commission or account credits if you sign up and rent, at no extra cost to you. Hourly rates and which GPUs are listed change all the time, so check the provider's page before you start. Editorial opinions are independent of the referral relationship; see how we make money.
How to read these numbers
Want a specific benchmark we don't have? Email Contact support and we'll prioritize it.