Your rig

M4 Max · 48 GB · 48 GB

48 GB unified · 546 GB/s · 36.0 GB addressable by the GPU
Edit hardware Export as JSON

Context length

KV cache at f16, 128K 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 Max · 48 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 Q8_0 and 128K. The ordering is editorial; the sizes and speeds are computed. How the shortlists work.

RecommendationModelWhyQuant · totalTok/s est.
Best overall Gemma 4 12B
Google · Apache 2.0
The 12 GB generalist: text, image and audio in, 140+ languages. Q8_0 · 20.8 GB 26 Runs great
Best for coding Qwen3.5 9B
Alibaba · Apache 2.0
The best coding model for 8–12 GB cards. Q8_0 · 14.1 GB 32 Runs great
Best reasoning gpt-oss 20B
OpenAI · Apache 2.0
Low/medium/high reasoning effort in 13 GB. MXFP4 · 14.4 GB 76 Runs great
Best for writing Gemma 4 12B
Google · Apache 2.0
The 12 GB writing pick; 140+ languages. Q8_0 · 20.8 GB 26 Runs great
Best vision Gemma 4 26B-A4B
Google · Apache 2.0
Vision at MoE speed in 18 GB. Q8_0 · 32.0 GB 38 Tight fit
Best for agents Muse Glimmer 30B
Meta · Apache 2.0
Distilled specifically for always-on local agent workflows. Q8_0 · 31.8 GB 10 Tight fit
Best translation Gemma 4 12B
Google · Apache 2.0
Broad language coverage on a 12 GB card. Q8_0 · 20.8 GB 26 Runs great
Fastest good model Gemma 4 26B-A4B
Google · Apache 2.0
3.8B active and a small file — the quickest Gemma 4 with vision. Q8_0 · 32.0 GB 38 Tight fit
Best long context Nemotron 3.5 Lightning 30B-A3B
NVIDIA · OpenMDW-1.1
262K, and a Mamba hybrid whose cache barely grows. Q8_0 · 32.6 GB
at 128K context
45 Tight fit

Everything that fits

Runs great

33 of 80 · fits with headroom for a longer prompt
ModelParamsQuantWeights +KV 128KTotal VRAM headroom Tok/sRuntime
gpt-oss 20B 20.9B MoE / 3.6B act MXFP4 10.8 GB 3.00 GB 14.4 GB
21.6 GB free
76 MLX Run
Phi-4 14B 14.7B Q8_0 14.5 GB 3.13 GB 18.3 GB
17.7 GB free
21 MLX Run
Gemma 3 12B 12.2B Q8_0 12.1 GB 8.31 GB 21.0 GB
15.0 GB free
25 MLX Run
Gemma 4 12B 12B Q8_0 11.9 GB 8.31 GB 20.8 GB
15.2 GB free
26 MLX Run
Qwen3.5 9B 9.65B Q8_0 9.5 GB 4.00 GB 14.1 GB
21.9 GB free
32 MLX Run
Ornith 1.5 9B 9.41B Q8_0 9.3 GB 4.00 GB 13.9 GB
22.1 GB free
33 MLX Run
Ministral 3 8B 8.92B Q8_0 8.8 GB 17.00 GB 26.4 GB
9.6 GB free
35 MLX Run
Granite 4.1 8B 8.79B Q8_0 8.7 GB 20.00 GB 29.3 GB
6.7 GB free
35 MLX Run
LFM2.5 8B-A1B 8.47B MoE / 1.5B act Q8_0 8.4 GB 1.46 GB 10.4 GB
25.6 GB free
96 MLX Run
Fara 7B 8.29B Q8_0 8.2 GB 6.84 GB 15.6 GB
20.4 GB free
37 MLX Run
Qwen3 8B 8.19B Q8_0 8.1 GB 18.00 GB 26.7 GB
9.3 GB free
38 MLX Run
Llama 3.1 8B Instruct 8.03B Q8_0 7.9 GB 16.00 GB 24.5 GB
11.5 GB free
38 MLX Run
Gemma 4 E4B 8.0B Q8_0 7.9 GB 1.78 GB 10.3 GB
25.7 GB free
39 MLX Run
Ling 3.0 Tiny 7.9B-A1.3B 7.9B MoE / 1.3B act Q8_0 7.8 GB 0.84 GB 9.3 GB
26.7 GB free
111 MLX Run
DeepSeek-R1-Distill-Qwen 7B 7.62B Q8_0 7.5 GB 7.00 GB 15.1 GB
20.9 GB free
40 MLX Run
Qwen2.5-Coder 7B 7.62B Q8_0 7.5 GB 7.00 GB 15.1 GB
20.9 GB free
40 MLX Run
Olmo 3 7B Instruct 7.3B Q8_0 7.2 GB 9.50 GB 17.3 GB
18.7 GB free
42 MLX Run
Mistral 7B Instruct v0.3 7.25B Q8_0 7.2 GB 4.00 GB 11.8 GB
24.2 GB free
43 MLX Run
Gemma 4 E2B 5.1B Q8_0 5.0 GB 0.89 GB 6.5 GB
29.5 GB free
60 MLX Run
Qwen3.5 4B 4.66B Q8_0 4.6 GB 4.00 GB 9.2 GB
26.8 GB free
66 MLX Run
Gemma 3 4B 4.3B Q8_0 4.3 GB 3.11 GB 8.0 GB
28.0 GB free
72 MLX Run
Qwen3 4B 4.02B Q8_0 4.0 GB 4.50 GB 9.1 GB
26.9 GB free
77 MLX Run
Ministral 3 3B 3.85B Q8_0 3.8 GB 13.00 GB 17.4 GB
18.6 GB free
80 MLX Run
Phi-4-mini 3.8B 3.84B Q8_0 3.8 GB 16.00 GB 20.4 GB
15.6 GB free
80 MLX Run
Granite 4.1 3B 3.4B Q8_0 3.4 GB 10.00 GB 14.0 GB
22.0 GB free
91 MLX Run
Llama 3.2 3B Instruct 3.21B Q8_0 3.2 GB 14.00 GB 17.8 GB
18.2 GB free
96 MLX Run
LFM2.5 2.6B 2.7B Q8_0 2.7 GB 2.00 GB 5.3 GB
30.7 GB free
114 MLX Run
Qwen3.5 2B 2.27B Q8_0 2.2 GB 1.50 GB 4.3 GB
31.7 GB free
136 MLX Run
Qwen3 1.7B 1.72B Q8_0 1.7 GB 3.50 GB 5.8 GB
30.2 GB free
179 MLX Run
Llama 3.2 1B Instruct 1.24B Q8_0 1.2 GB 4.00 GB 5.8 GB
30.2 GB free
249 MLX Run
Gemma 3 1B 1.0B Q8_0 1.0 GB 0.14 GB 1.7 GB
34.3 GB free
308 MLX Run
Qwen3.5 0.8B 0.87B Q8_0 0.9 GB 1.50 GB 3.0 GB
33.0 GB free
354 MLX Run
Qwen3 0.6B 0.6B Q8_0 0.6 GB 3.50 GB 4.7 GB
31.3 GB free
514 MLX Run

Runs — tight

6 models · fits at 128K, drop context or quant to go longer
ModelParamsQuantWeights +KV 128KTotal VRAM headroom Tok/sRuntime
Nemotron 3.5 Lightning 30B-A3B 31.6B MoE / 3.2B act Q8_0 31.3 GB 0.75 GB 32.6 GB
3.4 GB free
45 MLX Run
Muse Glimmer 30B 29.8B Q8_0 29.5 GB 1.70 GB 31.8 GB
4.2 GB free
10 MLX Run
Gemma 4 26B-A4B 26.5B MoE / 3.8B act Q8_0 26.2 GB 5.20 GB 32.0 GB
4.0 GB free
38 MLX Run
Qwen3 14B 14.8B Q8_0 14.6 GB 20.00 GB 35.2 GB
0.8 GB free
21 MLX Run
Ministral 3 14B 13.9B Q8_0 13.8 GB 20.00 GB 34.4 GB
1.6 GB free
22 MLX Run
Mistral NeMo 12B 12.2B Q8_0 12.1 GB 20.00 GB 32.7 GB
3.3 GB free
25 MLX Run

Won't fit in VRAM

41 models · some are possible with CPU offload
ModelParamsQuantWeights +KV 128KTotal Over budget Tok/sRuntime
Kimi K3 2.8T-A104B 2780B MoE / 104B act Q8_0 2750.9 GB 3.38 GB 2754.9 GB +2718.9 GB ~1.4 MLX Why
Qwen3.8 2.4T-A95B 2446B MoE / 95B act Q8_0 2420.4 GB 11.50 GB 2432.5 GB +2396.5 GB ~1.5 MLX Why
DeepSeek V4 Pro 1.6T-A49B 1650B MoE / 49B act Q8_0 1632.7 GB 8.58 GB 1641.9 GB +1605.9 GB ~2.9 MLX Why
Kimi K2.6 1T-A32B 1027B MoE / 32B act Q8_0 1016.2 GB 8.58 GB 1025.4 GB +989.4 GB ~4.5 MLX Why
GLM-5.2 744B-A40B 753B MoE / 40B act Q8_0 745.1 GB 10.97 GB 756.7 GB +720.7 GB ~3.6 MLX Why
GLM-5.3 744B-A40B 753B MoE / 40B act Q8_0 745.1 GB 10.97 GB 756.7 GB +720.7 GB ~3.6 MLX Why
DeepSeek-R1 671B 671B MoE / 37B act Q8_0 664.0 GB 8.58 GB 673.2 GB +637.2 GB ~3.9 MLX Why
DeepSeek V4.1 Flash 552B-A16B New 552B MoE / 16B act Q8_0 546.2 GB 0.57 GB 547.4 GB +511.4 GB ~9.0 DwarfStar Why
Qwen3.5 397B-A17B 403B MoE / 17B act Q8_0 398.8 GB 3.75 GB 403.1 GB +367.1 GB ~8.5 MLX Why
Ornith 1.5 397B-A17B 397B MoE / 17B act Q8_0 392.8 GB 3.75 GB 397.2 GB +361.2 GB ~8.5 MLX Why
GLM-5.3-Flash 320B-A18B 321B MoE / 18B act Q8_0 317.6 GB 1.38 GB 319.6 GB +283.6 GB ~8.0 MLX Why
DeepSeek V4 Flash 284B-A13B 284B MoE / 13B act Q8_0 281.0 GB 6.05 GB 287.7 GB +251.7 GB ~11 MLX Why
Qwen3 235B-A22B 235B MoE / 22B act Q8_0 232.5 GB 23.50 GB 256.6 GB +220.6 GB ~6.5 MLX Why
Qwen3.8-Flash-Next 180B-A6B 180B MoE / 6B act Q8_0 178.1 GB 3.00 GB 181.7 GB +145.7 GB ~24 MLX Why
Qwen3.5 122B-A10B 125B MoE / 10B act Q8_0 123.7 GB 3.00 GB 127.3 GB +91.3 GB ~14 MLX Why
Nemotron 3 Super 120B-A12B 124B MoE / 12B act Q8_0 122.7 GB 1.00 GB 124.3 GB +88.3 GB ~12 MLX Why
Ling 3.0 Flash 124B-A5B 124B MoE / 5.1B act Q8_0 122.7 GB 0.98 GB 124.3 GB +88.3 GB ~28 MLX Why
Mistral Small 4 119B-A6B 119B MoE / 6.5B act Q8_0 117.8 GB 2.81 GB 121.2 GB +85.2 GB ~22 MLX Why
gpt-oss 120B 117B MoE / 5.1B act MXFP4 60.5 GB 4.50 GB 65.6 GB +29.6 GB ~54 MLX Why
Llama 4 Scout 109B-A17B 109B MoE / 17B act Q8_0 107.9 GB 7.13 GB 115.6 GB +79.6 GB ~8.5 MLX Why
Llama 3.3 70B Instruct 70.6B Q8_0 69.9 GB 40.00 GB 110.5 GB +74.5 GB ~4.4 MLX Why
DeepSeek-R1-Distill-Llama 70B 70.6B Q8_0 69.9 GB 40.00 GB 110.5 GB +74.5 GB ~4.4 MLX Why
Qwen3.6 35B-A3B 35.9B MoE / 3.3B act Q8_0 35.5 GB 2.50 GB 38.6 GB +2.6 GB ~44 MLX Why
Ornith 1.5 35B-A3B 35.9B MoE / 3.3B act Q8_0 35.5 GB 2.50 GB 38.6 GB +2.6 GB ~44 MLX Why
LLM-jp 4 33B Thinking 33.2B Q8_0 32.9 GB 16.00 GB 49.5 GB +13.5 GB ~9.3 MLX Why
Qwen3 32B 32.8B Q8_0 32.5 GB 32.00 GB 65.1 GB +29.1 GB ~9.4 MLX Why
DeepSeek-R1-Distill-Qwen 32B 32.8B Q8_0 32.5 GB 32.00 GB 65.1 GB +29.1 GB ~9.4 MLX Why
Qwen2.5-Coder 32B 32.8B Q8_0 32.5 GB 32.00 GB 65.1 GB +29.1 GB ~9.4 MLX Why
Olmo 3.1 32B Instruct 32.2B Q8_0 31.9 GB 4.75 GB 37.2 GB +1.2 GB ~9.6 MLX Why
Gemma 4 31B 31.3B Q8_0 31.0 GB 20.78 GB 52.4 GB +16.4 GB ~9.9 MLX Why
GLM-4.7-Flash 30B-A3B 31.2B MoE / 3B act Q8_0 30.9 GB 6.61 GB 38.1 GB +2.1 GB ~48 MLX Why
Qwen3 30B-A3B 30.5B MoE / 3.3B act Q8_0 30.2 GB 12.00 GB 42.8 GB +6.8 GB ~44 MLX Why
Qwen3 Coder 30B-A3B 30.5B MoE / 3.3B act Q8_0 30.2 GB 12.00 GB 42.8 GB +6.8 GB ~44 MLX Why
Granite 4.1 30B 28.9B Q8_0 28.6 GB 32.00 GB 61.2 GB +25.2 GB ~11 MLX Why
Qwen3.8 27B 27.8B Q8_0 27.5 GB 8.00 GB 36.1 GB +0.1 GB ~11 MLX Why
Qwen3.6 27B 27.8B Q8_0 27.5 GB 8.00 GB 36.1 GB +0.1 GB ~11 MLX Why
Gemma 3 27B 27.4B Q8_0 27.1 GB 10.41 GB 38.1 GB +2.1 GB ~11 MLX Why
Devstral Small 2 24B 24B Q8_0 23.7 GB 20.00 GB 44.3 GB +8.3 GB ~13 MLX Why
Mistral Small 3.2 24B 23.6B Q8_0 23.4 GB 20.00 GB 44.0 GB +8.0 GB ~13 MLX Why
DeepSeek-R1-Distill-Qwen 14B 14.8B Q8_0 14.6 GB 24.00 GB 39.2 GB +3.2 GB ~21 MLX Why
Qwen2.5-Coder 14B 14.8B Q8_0 14.6 GB 24.00 GB 39.2 GB +3.2 GB ~21 MLX Why

Totals = quantised weights + f16 KV cache at 128K tokens + 0.6 GB runtime overhead, against 36.0 GB usable unified memory. Tokens per second are estimated from 546 GB/s peak memory bandwidth at batch 1 — not measured. Read the method.


Same memory, different speed

These machines have the same 36.0 GB usable, so they run exactly the models above. Generation speed scales with memory bandwidth.

DeviceBandwidthSpeed vs yours
M4 Pro · 48 GB
M4 Pro · 2024
273 GB/s 0.50×

Best LLM for a 48 GB Mac, with every machine in the class →

What more memory unlocks

At Q4_K_M and 8K context this machine fits 58 of 80 models. Each step up, and the largest models it adds:

Every device ranked by what it can run →