What local LLMs run on M4 16GB?

With 16 GB unified memory (about 10.7 GB usable by the GPU), these are the best open models that fit. The top pick is DeepSeek R1 Distill Qwen 14B at Q4_K_M (~10.3 GB, ~11 tok/s): ollama run deepseek-r1:14b.

Best for coding: DeepSeek Coder V2 Lite 16B (MoE) (deepseek-coder-v2:16b), Qwen 2.5 Coder 14B (qwen2.5-coder:14b), Qwen 2.5 Coder 7B (qwen2.5-coder:7b-instruct-q8_0).

Runtimes available for this hardware: Ollama, LM Studio, llama.cpp, MLX.

Open the interactive checker for M4 16GB

#	Model	Quant	Memory needed	Fit	Speed	Command
1	DeepSeek R1 Distill Qwen 14B	Q4_K_M	10.3 GB	fits in fast memory	~11 tok/s	`ollama run deepseek-r1:14b`
2	Phi-4 14B	Q4_K_M	10.4 GB	fits in fast memory	~11 tok/s	`ollama run phi4:14b`
3	Qwen3 14B	Q4_K_M	10.6 GB	fits in fast memory	~10 tok/s	`ollama run qwen3:14b`
4	Mistral Nemo 12B	Q5_K_M	9.9 GB	fits in fast memory	~11 tok/s	`ollama run mistral-nemo:12b-instruct-2407-q5_K_M`
5	Gemma 3 12B	Q4_K_M	9.4 GB	fits in fast memory	~12 tok/s	`ollama run gemma3:12b`
6	Llama 3.2 Vision 11B	Q4_K_M	8.9 GB	fits in fast memory	~12 tok/s	`ollama run llama3.2-vision:11b`
7	Nimble v3 9B (Bespoke Labs, decision)	Q8_0	10.6 GB	fits in fast memory	~10 tok/s	`ollama run nimble-v3:9b-q8_0`
8	Qwen3 8B	Q8_0	9.9 GB	fits in fast memory	~11 tok/s	`ollama run qwen3:8b-q8_0`
9	DeepSeek R1 0528 Qwen3 8B	Q8_0	9.9 GB	fits in fast memory	~11 tok/s	`ollama run deepseek-r1:8b-0528-qwen3-q8_0`
10	DeepSeek R1 Distill Qwen 7B	Q8_0	9.1 GB	fits in fast memory	~12 tok/s	`ollama run deepseek-r1:7b-qwen-distill-q8_0`
11	Qwen 2.5 VL 7B	Q8_0	10.4 GB	fits in fast memory	~10 tok/s	`ollama run qwen2.5vl:7b-q8_0`
12	MiniCPM-V 2.6 8B	Q8_0	10.1 GB	fits in fast memory	~11 tok/s	`ollama run minicpm-v:8b-2.6-q8_0`

Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-14.

A project by OnDevice Labs · llms.txt