Your rig

M4 · 16 GB · 16 GB

16 GB unified · 120 GB/s · 10.7 GB addressable by the GPU
Edit hardware Export as JSON

Context length

KV cache at f16, 8K tokens

Quality floor
Q8_0 — near-lossless Q6_K Q5_K_M Q4_K_M — recommended Q3_K_M Q2_K — damaged
KV cache type
Runtime
Ollamallama.cppLM StudioMLX

4 of 5 runtimes drive this device.

Allow CPU offload Include partial-offload fits

Offloaded layers run at 60 GB/s host bandwidth — expect single-digit tokens per second.

Which one should I actually run?

Best models for your M4 · 16 GB

Updated 8 Oct 2026 1 new this month

What do you want to do? Each row is the best-ranked model on that use case's shortlist that fits at Q4_K_M. The ordering is editorial; the sizes and speeds are computed. How the shortlists work.

RecommendationModelWhyQuant · totalTok/s est.
Best long context
Your pick
Qwen3.5 4B
Alibaba · Apache 2.0
262K on 8 GB. Q4_K_M · 7.2 GB
at 128K context
26 Runs great
Best overall Gemma 4 12B
Google · Apache 2.0
The 12 GB generalist: text, image and audio in, 140+ languages. Q4_K_M · 8.2 GB 9.9 Runs great
Best for coding Qwen3.5 9B
Alibaba · Apache 2.0
The best coding model for 8–12 GB cards. Q4_K_M · 6.3 GB 12 Runs great
Best reasoning Qwen3.5 9B
Alibaba · Apache 2.0
Thinking mode on a 12 GB card — ahead of the 2025 R1 distills. Q4_K_M · 6.3 GB 12 Runs great
Best for writing Gemma 4 12B
Google · Apache 2.0
The 12 GB writing pick; 140+ languages. Q4_K_M · 8.2 GB 9.9 Runs great
Best vision Gemma 4 12B
Google · Apache 2.0
Image and audio understanding on 12 GB. Q4_K_M · 8.2 GB 9.9 Runs great
Best for agents Granite 4.1 8B
IBM · Apache 2.0
Enterprise-grade tool calling in a dense 8B, no thinking overhead. Q4_K_M · 6.8 GB 14 Runs great
Best translation Gemma 4 12B
Google · Apache 2.0
Broad language coverage on a 12 GB card. Q4_K_M · 8.2 GB 9.9 Runs great
Fastest good model LFM2.5 8B-A1B
Liquid AI · LFM Open License v1.0
1.5B active; designed for laptops without a GPU. Q4_K_M · 5.5 GB 37 Runs great

Everything that fits

Runs great

32 of 80 · fits with headroom for a longer prompt
ModelParamsQuantWeights +KV 8KTotal VRAM headroom Tok/sRuntime
Gemma 3 12B 12.2B Q4_K_M 6.9 GB 0.81 GB 8.3 GB
2.4 GB free
9.8 MLX Run
Mistral NeMo 12B 12.2B Q4_K_M 6.9 GB 1.25 GB 8.7 GB
2.0 GB free
9.8 MLX Run
Gemma 4 12B 12B Q4_K_M 6.7 GB 0.81 GB 8.2 GB
2.5 GB free
9.9 MLX Run
Qwen3.5 9B 9.65B Q4_K_M 5.4 GB 0.25 GB 6.3 GB
4.4 GB free
12 MLX Run
Ornith 1.5 9B 9.41B Q4_K_M 5.3 GB 0.25 GB 6.1 GB
4.6 GB free
13 MLX Run
Ministral 3 8B 8.92B Q4_K_M 5.0 GB 1.06 GB 6.7 GB
4.0 GB free
13 MLX Run
Granite 4.1 8B 8.79B Q4_K_M 4.9 GB 1.25 GB 6.8 GB
3.9 GB free
14 MLX Run
LFM2.5 8B-A1B 8.47B MoE / 1.5B act Q4_K_M 4.8 GB 0.09 GB 5.5 GB
5.2 GB free
37 MLX Run
Fara 7B 8.29B Q4_K_M 4.7 GB 0.44 GB 5.7 GB
5.0 GB free
14 MLX Run
Qwen3 8B 8.19B Q4_K_M 4.6 GB 1.13 GB 6.3 GB
4.4 GB free
15 MLX Run
Llama 3.1 8B Instruct 8.03B Q4_K_M 4.5 GB 1.00 GB 6.1 GB
4.6 GB free
15 MLX Run
Gemma 4 E4B 8.0B Q4_K_M 4.5 GB 0.14 GB 5.2 GB
5.5 GB free
15 MLX Run
Ling 3.0 Tiny 7.9B-A1.3B 7.9B MoE / 1.3B act Q4_K_M 4.4 GB 0.05 GB 5.1 GB
5.6 GB free
43 MLX Run
DeepSeek-R1-Distill-Qwen 7B 7.62B Q4_K_M 4.3 GB 0.44 GB 5.3 GB
5.4 GB free
16 MLX Run
Qwen2.5-Coder 7B 7.62B Q4_K_M 4.3 GB 0.44 GB 5.3 GB
5.4 GB free
16 MLX Run
Olmo 3 7B Instruct 7.3B Q4_K_M 4.1 GB 2.50 GB 7.2 GB
3.5 GB free
16 MLX Run
Mistral 7B Instruct v0.3 7.25B Q4_K_M 4.1 GB 1.00 GB 5.7 GB
5.0 GB free
16 MLX Run
Gemma 4 E2B 5.1B Q4_K_M 2.9 GB 0.07 GB 3.5 GB
7.2 GB free
23 MLX Run
Qwen3.5 4B 4.66B Q4_K_M 2.6 GB 0.25 GB 3.5 GB
7.2 GB free
26 MLX Run
Gemma 3 4B 4.3B Q4_K_M 2.4 GB 0.30 GB 3.3 GB
7.4 GB free
28 MLX Run
Qwen3 4B 4.02B Q4_K_M 2.3 GB 1.13 GB 4.0 GB
6.7 GB free
30 MLX Run
Ministral 3 3B 3.85B Q4_K_M 2.2 GB 0.81 GB 3.6 GB
7.1 GB free
31 MLX Run
Phi-4-mini 3.8B 3.84B Q4_K_M 2.2 GB 1.00 GB 3.8 GB
6.9 GB free
31 MLX Run
Granite 4.1 3B 3.4B Q4_K_M 1.9 GB 0.63 GB 3.1 GB
7.6 GB free
35 MLX Run
Llama 3.2 3B Instruct 3.21B Q4_K_M 1.8 GB 0.88 GB 3.3 GB
7.4 GB free
37 MLX Run
LFM2.5 2.6B 2.7B Q4_K_M 1.5 GB 0.13 GB 2.2 GB
8.5 GB free
44 MLX Run
Qwen3.5 2B 2.27B Q4_K_M 1.3 GB 0.09 GB 2.0 GB
8.7 GB free
53 MLX Run
Qwen3 1.7B 1.72B Q4_K_M 1.0 GB 0.88 GB 2.4 GB
8.3 GB free
69 MLX Run
Llama 3.2 1B Instruct 1.24B Q4_K_M 0.7 GB 0.25 GB 1.5 GB
9.2 GB free
96 MLX Run
Gemma 3 1B 1.0B Q4_K_M 0.6 GB 0.04 GB 1.2 GB
9.5 GB free
119 MLX Run
Qwen3.5 0.8B 0.87B Q4_K_M 0.5 GB 0.09 GB 1.2 GB
9.5 GB free
137 MLX Run
Qwen3 0.6B 0.6B Q4_K_M 0.3 GB 0.88 GB 1.8 GB
8.9 GB free
199 MLX Run

Runs — tight

5 models · fits at 8K, drop context or quant to go longer
ModelParamsQuantWeights +KV 8KTotal VRAM headroom Tok/sRuntime
Qwen3 14B 14.8B Q4_K_M 8.3 GB 1.25 GB 10.2 GB
0.5 GB free
8.1 MLX Run
DeepSeek-R1-Distill-Qwen 14B 14.8B Q4_K_M 8.3 GB 1.50 GB 10.4 GB
0.3 GB free
8.1 MLX Run
Qwen2.5-Coder 14B 14.8B Q4_K_M 8.3 GB 1.50 GB 10.4 GB
0.3 GB free
8.1 MLX Run
Phi-4 14B 14.7B Q4_K_M 8.3 GB 1.56 GB 10.4 GB
0.3 GB free
8.1 MLX Run
Ministral 3 14B 13.9B Q4_K_M 7.8 GB 1.25 GB 9.7 GB
1.0 GB free
8.6 MLX Run

Won't fit in VRAM

43 models · some are possible with CPU offload
ModelParamsQuantWeights +KV 8KTotal Over budget Tok/sRuntime
Kimi K3 2.8T-A104B 2780B MoE / 104B act Q4_K_M 1563.2 GB 0.21 GB 1564.0 GB +1553.3 GB ~0.5 MLX Why
Qwen3.8 2.4T-A95B 2446B MoE / 95B act Q4_K_M 1375.4 GB 0.72 GB 1376.7 GB +1366.0 GB ~0.6 MLX Why
DeepSeek V4 Pro 1.6T-A49B 1650B MoE / 49B act Q4_K_M 927.8 GB 0.54 GB 928.9 GB +918.2 GB ~1.1 MLX Why
Kimi K2.6 1T-A32B 1027B MoE / 32B act Q4_K_M 577.5 GB 0.54 GB 578.6 GB +567.9 GB ~1.7 MLX Why
GLM-5.2 744B-A40B 753B MoE / 40B act Q4_K_M 423.4 GB 0.69 GB 424.7 GB +414.0 GB ~1.4 MLX Why
GLM-5.3 744B-A40B 753B MoE / 40B act Q4_K_M 423.4 GB 0.69 GB 424.7 GB +414.0 GB ~1.4 MLX Why
DeepSeek-R1 671B 671B MoE / 37B act Q4_K_M 377.3 GB 0.54 GB 378.4 GB +367.7 GB ~1.5 MLX Why
DeepSeek V4.1 Flash 552B-A16B New 552B MoE / 16B act Q4_K_M 310.4 GB 0.04 GB 311.0 GB +300.3 GB ~3.5 DwarfStar Why
Qwen3.5 397B-A17B 403B MoE / 17B act Q4_K_M 226.6 GB 0.23 GB 227.4 GB +216.7 GB ~3.3 MLX Why
Ornith 1.5 397B-A17B 397B MoE / 17B act Q4_K_M 223.2 GB 0.23 GB 224.1 GB +213.4 GB ~3.3 MLX Why
GLM-5.3-Flash 320B-A18B 321B MoE / 18B act Q4_K_M 180.5 GB 0.09 GB 181.2 GB +170.5 GB ~3.1 MLX Why
DeepSeek V4 Flash 284B-A13B 284B MoE / 13B act Q4_K_M 159.7 GB 0.38 GB 160.7 GB +150.0 GB ~4.3 MLX Why
Qwen3 235B-A22B 235B MoE / 22B act Q4_K_M 132.1 GB 1.47 GB 134.2 GB +123.5 GB ~2.5 MLX Why
Qwen3.8-Flash-Next 180B-A6B 180B MoE / 6B act Q4_K_M 101.2 GB 0.19 GB 102.0 GB +91.3 GB ~9.3 MLX Why
Qwen3.5 122B-A10B 125B MoE / 10B act Q4_K_M 70.3 GB 0.19 GB 71.1 GB +60.4 GB ~5.6 MLX Why
Nemotron 3 Super 120B-A12B 124B MoE / 12B act Q4_K_M 69.7 GB 0.06 GB 70.4 GB +59.7 GB ~4.6 MLX Why
Ling 3.0 Flash 124B-A5B 124B MoE / 5.1B act Q4_K_M 69.7 GB 0.06 GB 70.4 GB +59.7 GB ~11 MLX Why
Mistral Small 4 119B-A6B 119B MoE / 6.5B act Q4_K_M 66.9 GB 0.18 GB 67.7 GB +57.0 GB ~8.6 MLX Why
gpt-oss 120B 117B MoE / 5.1B act MXFP4 60.5 GB 0.29 GB 61.4 GB +50.7 GB ~12 MLX Why
Llama 4 Scout 109B-A17B 109B MoE / 17B act Q4_K_M 61.3 GB 1.50 GB 63.4 GB +52.7 GB ~3.3 MLX Why
Llama 3.3 70B Instruct 70.6B Q4_K_M 39.7 GB 2.50 GB 42.8 GB +32.1 GB ~1.7 MLX Why
DeepSeek-R1-Distill-Llama 70B 70.6B Q4_K_M 39.7 GB 2.50 GB 42.8 GB +32.1 GB ~1.7 MLX Why
Qwen3.6 35B-A3B 35.9B MoE / 3.3B act Q4_K_M 20.2 GB 0.16 GB 20.9 GB +10.2 GB ~17 MLX Why
Ornith 1.5 35B-A3B 35.9B MoE / 3.3B act Q4_K_M 20.2 GB 0.16 GB 20.9 GB +10.2 GB ~17 MLX Why
LLM-jp 4 33B Thinking 33.2B Q4_K_M 18.7 GB 2.00 GB 21.3 GB +10.6 GB ~3.6 MLX Why
Qwen3 32B 32.8B Q4_K_M 18.4 GB 2.00 GB 21.0 GB +10.3 GB ~3.6 MLX Why
DeepSeek-R1-Distill-Qwen 32B 32.8B Q4_K_M 18.4 GB 2.00 GB 21.0 GB +10.3 GB ~3.6 MLX Why
Qwen2.5-Coder 32B 32.8B Q4_K_M 18.4 GB 2.00 GB 21.0 GB +10.3 GB ~3.6 MLX Why
Olmo 3.1 32B Instruct 32.2B Q4_K_M 18.1 GB 1.25 GB 20.0 GB +9.3 GB ~3.7 MLX Why
Nemotron 3.5 Lightning 30B-A3B 31.6B MoE / 3.2B act Q4_K_M 17.8 GB 0.05 GB 18.4 GB +7.7 GB ~17 MLX Why
Gemma 4 31B 31.3B Q4_K_M 17.6 GB 2.03 GB 20.2 GB +9.5 GB ~3.8 MLX Why
GLM-4.7-Flash 30B-A3B 31.2B MoE / 3B act Q4_K_M 17.5 GB 0.41 GB 18.6 GB +7.9 GB ~19 MLX Why
Qwen3 30B-A3B 30.5B MoE / 3.3B act Q4_K_M 17.1 GB 0.75 GB 18.5 GB +7.8 GB ~17 MLX Why
Qwen3 Coder 30B-A3B 30.5B MoE / 3.3B act Q4_K_M 17.1 GB 0.75 GB 18.5 GB +7.8 GB ~17 MLX Why
Muse Glimmer 30B 29.8B Q4_K_M 16.8 GB 0.18 GB 17.5 GB +6.8 GB ~4.0 MLX Why
Granite 4.1 30B 28.9B Q4_K_M 16.3 GB 2.00 GB 18.9 GB +8.2 GB ~4.1 MLX Why
Qwen3.8 27B 27.8B Q4_K_M 15.6 GB 0.50 GB 16.7 GB +6.0 GB ~4.3 MLX Why
Qwen3.6 27B 27.8B Q4_K_M 15.6 GB 0.50 GB 16.7 GB +6.0 GB ~4.3 MLX Why
Gemma 3 27B 27.4B Q4_K_M 15.4 GB 1.03 GB 17.0 GB +6.3 GB ~4.4 MLX Why
Gemma 4 26B-A4B 26.5B MoE / 3.8B act Q4_K_M 14.9 GB 0.51 GB 16.0 GB +5.3 GB ~15 MLX Why
Devstral Small 2 24B 24B Q4_K_M 13.5 GB 1.25 GB 15.3 GB +4.6 GB ~5.0 MLX Why
Mistral Small 3.2 24B 23.6B Q4_K_M 13.3 GB 1.25 GB 15.1 GB +4.4 GB ~5.1 MLX Why
gpt-oss 20B 20.9B MoE / 3.6B act MXFP4 10.8 GB 0.19 GB 11.6 GB +0.9 GB ~17 MLX Why

Totals = quantised weights + f16 KV cache at 8K tokens + 0.6 GB runtime overhead, against 10.7 GB usable unified memory. Tokens per second are estimated from 120 GB/s peak memory bandwidth at batch 1 — not measured. Read the method.


Same memory, different speed

These machines have the same 10.7 GB usable, so they run exactly the models above. Generation speed scales with memory bandwidth.

DeviceBandwidthSpeed vs yours
M1 · 16 GB
M1 · 2020
68 GB/s 0.57×

Best LLM for a 16 GB Mac, with every machine in the class →

What more memory unlocks

At Q4_K_M and 8K context this machine fits 37 of 80 models. Each step up, and the largest models it adds:

Every device ranked by what it can run →