Can Mac mini M4 Pro 24 GB run Kolibri-1 (78B-A3.5B)?
24 GB unified · 273 GB/s memory bandwidth · Kolibri-1 (78B-A3.5B) has 78.1B parameters (3.46B active per token).
Short answer
No. It doesn't fit in memory.
Needs ≈46 GB but only ≈17 GB is available after the OS, other apps and Open WebUI.
Why it doesn’t fit
Even its smallest practical version (Q4_K_M) needs about 46 GB for the model, its context memory and the runtime. After macOS (≈3.5 GB), a browser with a few tabs (≈3.0 GB) and Open WebUI (≈1.0 GB), Mac mini M4 Pro 24 GB has about 17 GB left, so it’s 29 GB short.
Ways to make it work
Close everything else
Not enough on its own: with every other app closed it’s still about 26 GB short.
Pick a smaller model from the same family
No smaller Kolibri model in our catalog runs well here. See the other models below.
Get more memory
The same chip with 64 GB handles it: Excellent.
Check your exact setup
These pages assume typical settings. A shorter context or a lighter chat app can change the answer.
Check Mac mini M4 Pro 24 GB
The same applies to every workload we rate (chat, coding questions, coding in a repository, agentic coding and long documents): they all need at least as much memory as chat.
Other models for Mac mini M4 Pro 24 GB
The best-rated models for chat on this machine.
- gpt-oss-20b60–80 tok/sExcellent
- MiniCPM5 2B60–85 tok/sExcellent
- Qwen3.5 4B40–55 tok/sExcellent
- Qwen3.5 9B20–30 tok/sExcellent
- Gemma 4 12B25–35 tok/sExcellent
- GLM-4.6V-Flash (10B)30–40 tok/sExcellent
Cheapest machines for Kolibri-1 (78B-A3.5B)
Hardware rated comfortable or better for chat, lowest price first.
- Desktop PC · Ryzen 9 9950X (no GPU) · 64 GB DDR5$1,400 · 13–20 tok/sComfortable
- Ryzen AI Max+ 395 mini PC / Framework Desktop · 64 GB$1,600 · 65–90 tok/sExcellent
- Ryzen AI Max+ 395 mini PC / Framework Desktop · 128 GB$2,000 · 40–55 tok/sExcellent
- Mac mini M4 Pro 64 GB$2,199 · 60–80 tok/sExcellent
- Mac Studio M1 Max 64 GB$2,399 · 80–110 tok/sExcellent
- Mac mini M5 Pro 64 GB$2,499 · 55–90 tok/sExcellent
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.