VOOZH about

URL: https://willitrunai.com/my-results/m4-max-36gb

⇱ MacBook Pro M4 Max 36GB AI Tier List: 305 models ranked | WillItRunAI


AI Compatibility Tier List

MacBook Pro M4 Max 36GB

374 AI models ranked by how well they run on your 36 GB unified memory hardware. Best match: Qwen3-Coder 30B A3B Instruct (score: 94).

374

Models tested

305

Can run

36 GB

unified memory

39

Best tok/s

Tier Distribution

S 21
A 59
B 38
C 187
F 69
S: 21A: 59B: 38C: 187F: 69
S

S Tier -- Perfect Fit (21 models)

These models run flawlessly with headroom to spare.

👁 Alibaba
Qwen3-Coder 30B A3B Instruct
S94
30.5B24.9 GB39 tok/s28K ctx
moe
👁 Alibaba
Qwen3-VL 30B A3B Instruct
S93
30B24.6 GB40 tok/s31K ctx
moe
👁 OpenAI
GPT-OSS 20B
S93
21B20.0 GB50 tok/s55K ctx
moe
👁 Alibaba
Qwen 3.5 27B
S93
27B24.4 GB29 tok/s24K ctx
dense
👁 Alibaba
Qwen 3 14B
S92
14B15.8 GB33 tok/s83K ctx
dense
👁 Alibaba
Qwen 3.5 9B
S92
9B12.5 GB55 tok/s114K ctx
dense
👁 Microsoft
Phi-4-reasoning-plus 14B
S92
14.7B16.8 GB32 tok/s33K ctx
dense
👁 Mistral
Devstral Small 2 24B Instruct
S92
24B21.9 GB35 tok/s43K ctx
dense
👁 Alibaba
Qwen 3.6 27B
S92
27B22.2 GB22 tok/s76K ctx
+1dense
👁 Mistral
Magistral Small 2507
S92
24B21.9 GB29 tok/s43K ctx
dense
👁 Alibaba
Qwen 3 30B A3B
S92
30.5B24.9 GB39 tok/s28K ctx
moe
👁 NVIDIA
Nemotron 3 Nano 30B
S90
30B25.5 GB30 tok/s19K ctx
dense
👁 Alibaba
Qwen 3 8B
S90
8B11.9 GB62 tok/s118K ctx
dense
👁 Mistral
Devstral Small 1.1
S90
24B21.9 GB35 tok/s43K ctx
dense
👁 NVIDIA
Nemotron Cascade 2 30B A3B
S89
30B26.0 GB40 tok/s15K ctx
moe
👁 Alibaba
Qwen 3.5 4B
S88
4B9.4 GB56 tok/s131K ctx
dense
👁 Mistral
Codestral 2 25.08
S87
22B20.6 GB28 tok/s51K ctx
dense
👁 Mistral
Ministral 3 14B
S87
14B15.8 GB33 tok/s83K ctx
multimodal
S87
19B18.8 GB31 tok/s8K ctx
dense
👁 Google
Gemma 4 26B A4B
S86
25.2B23.8 GB42 tok/s25K ctx
moe
👁 NVIDIA
Nemotron Nano 8B
S86
8B11.6 GB67 tok/s131K ctx
dense
A

A Tier -- Great Fit (59 models)

Excellent performance with comfortable memory margins.

👁 Microsoft
Phi-4 Mini Reasoning 4B
A84
3.8B8.6 GB53 tok/s131K ctx
dense
👁 Mistral
Mistral Small 3.2 24B
A84
24B21.9 GB29 tok/s43K ctx
vision
👁 DeepSeek
DeepSeek Coder V2 16B
A84
16B17.8 GB63 tok/s55K ctx
moe
A84
14B16.4 GB33 tok/s16K ctx
dense
A83
24B21.9 GB29 tok/s43K ctx
dense
👁 IBM
Granite Code 20B
A83
20B20.2 GB31 tok/s8K ctx
dense
👁 Alibaba
Qwen 2.5 14B
A83
14B16.3 GB33 tok/s69K ctx
dense
👁 InternLM
InternVL2 8B
A83
8B11.6 GB62 tok/s8K ctx
dense
👁 Alibaba
Qwen 3.6 35B A3B
A82
35B30.2 GB29 tok/s4K ctx
+1moe
👁 Alibaba
Qwen 3.5 35B A3B
A82
35B27.6 GB35 tok/s4K ctx
moe
👁 Mistral
Mistral Small 24B
A82
24B21.9 GB29 tok/s33K ctx
dense
👁 IBM
Granite 4.1 30B
A82
30B27.0 GB26 tok/s12K ctx
dense
👁 Google
Gemma 3 12B
A82
12B17.0 GB28 tok/s45K ctx
dense
👁 Mistral AI
Mistral Small 3.1 24B
A81
24B21.9 GB29 tok/s43K ctx
dense
👁 Mistral
Ministral 3 8B
A81
8B11.9 GB62 tok/s118K ctx
multimodal
👁 OpenBMB
MiniCPM-V 2.6 8B
A81
8B11.6 GB62 tok/s2K ctx
dense
👁 NVIDIA
Nemotron Nano 9B v2
A80
9B12.7 GB60 tok/s103K ctx
dense
👁 Alibaba
Qwen 3 32B
A80
32B28.2 GB23 tok/s7K ctx
dense
👁 Defog
SQLCoder 7B
A80
7B11.0 GB71 tok/s8K ctx
dense
👁 Alibaba
Qwen 3 4B
A79
4B9.4 GB56 tok/s33K ctx
dense
👁 Alibaba
Qwen 2.5 VL 7B
A79
7B9.9 GB72 tok/s33K ctx
dense
👁 AllenAI
OLMo 2 13B
A78
13B15.2 GB35 tok/s33K ctx
dense
👁 Meta
CodeLlama 7B Instruct
A78
7B16.9 GB66 tok/s16K ctx
dense
👁 Magistral AI
Magistral 7B
A78
7B11.0 GB71 tok/s8K ctx
dense
👁 BigCode
StarCoder 7B
A78
7B16.4 GB66 tok/s8K ctx
dense
👁 Jina AI
Jina Embeddings v3
A78
0.57B7.9 GB8 tok/s8K ctx
dense
👁 Tsinghua/Zhipu
CodeGeeX 4 9B
A77
9B10.9 GB56 tok/s131K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 14B
A77
14B16.3 GB33 tok/s33K ctx
dense
👁 Google
Gemma 4 E4B
A76
8B10.9 GB47 tok/s128K ctx
dense
👁 BAAI
BGE M3
A76
0.57B7.1 GB8 tok/s8K ctx
dense
👁 Meta
CodeLlama 13B Instruct
A76
13B24.9 GB33 tok/s16K ctx
dense
👁 InternLM
InternLM 7B
A75
7B16.9 GB66 tok/s8K ctx
dense
👁 InternLM
InternLM Chat 7B
A75
7B16.9 GB66 tok/s8K ctx
dense
👁 IBM
Granite Code 8B
A75
8B11.6 GB62 tok/s8K ctx
dense
👁 DevStral AI
DevStral 7B
A75
7B11.0 GB71 tok/s8K ctx
dense
👁 Alibaba
Qwen 2.5 7B
A75
7B9.9 GB72 tok/s131K ctx
dense
👁 IBM
Granite 4.1 8B
A75
8B12.1 GB62 tok/s107K ctx
dense
👁 Mistral AI
Pixtral 12B
A75
12B14.5 GB38 tok/s91K ctx
dense
👁 LG AI
EXAONE 4.0 32B
A74
32B28.2 GB23 tok/s7K ctx
dense
👁 Mistral AI
Codestral Mamba 7B
A74
7B9.5 GB70 tok/s262K ctx
state-space
👁 LLaVA
LLaVA 1.6 13B
A74
13B24.9 GB33 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 32B
A73
32B28.2 GB23 tok/s7K ctx
dense
👁 LLaVA
LLaVA 1.5 7B
A73
7B16.9 GB66 tok/s4K ctx
dense
👁 Nous Research
Nous Hermes 1.0
A73
9B22.5 GB51 tok/s16K ctx
dense
👁 Mixedbread AI
mxbai Embed Large
A72
0.34B6.9 GB5 tok/s1K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 3B
A72
3B8.8 GB42 tok/s131K ctx
dense
👁 Allen AI
OLMo 2 32B
A72
32B28.2 GB23 tok/s4K ctx
dense
A72
7B12.0 GB71 tok/s8K ctx
dense
👁 Snowflake
Snowflake Arctic Embed L
A72
0.34B6.9 GB5 tok/s1K ctx
dense
👁 Meta
Llama 3.1 8B
A71
8B11.6 GB62 tok/s128K ctx
dense
👁 WizardLM
WizardLM 13B
A71
13B24.9 GB33 tok/s8K ctx
dense
👁 LMSYS
Vicuna 13B
A71
13B24.9 GB33 tok/s4K ctx
dense
👁 MosaicML
MPT-7B-Instruct
A71
7B16.9 GB66 tok/s8K ctx
dense
👁 Baichuan
Baichuan 7B
A71
7B16.9 GB66 tok/s8K ctx
dense
👁 Google
Gemma 4 E2B
A71
5.1B8.4 GB71 tok/s128K ctx
dense
👁 WizardLM
WizardMath 7B
A71
7B11.0 GB71 tok/s4K ctx
dense
👁 Allen AI
OLMo 2 7B
A71
7B11.0 GB71 tok/s4K ctx
dense
👁 Zhipu
GLM-4 9B
A71
9B10.9 GB56 tok/s128K ctx
dense
👁 Nous Research
Nous Dolphin 13B
A71
13B26.4 GB27 tok/s15K ctx
dense
B

B Tier -- Good Fit (38 models)

Solid performance, may need quantization or reduced context.

👁 BAAI
BGE Large EN v1.5
B70
0.34B6.9 GB5 tok/s1K ctx
dense
👁 Google
Gemma 3 4B
B69
4B9.3 GB56 tok/s128K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 7B
B69
7B9.9 GB72 tok/s131K ctx
dense
👁 Microsoft
Phi 4 Mini 4B
B68
4B8.7 GB56 tok/s128K ctx
dense
👁 Microsoft
Phi 3 Mini 3.8B
B68
3.8B13.0 GB53 tok/s51K ctx
dense
👁 Mistral
Ministral 3 3B
B68
3B7.4 GB42 tok/s262K ctx
multimodal
👁 Nomic AI
Nomic Embed Text v1.5
B67
0.14B5.6 GB2 tok/s8K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 32B
B67
32B28.2 GB23 tok/s7K ctx
dense
👁 Microsoft
Phi 3.5 Mini 4B
B67
4B13.1 GB56 tok/s51K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 8B
B67
8B11.6 GB62 tok/s33K ctx
dense
👁 Alibaba
Qwen 3.5 2B
B67
2B7.7 GB28 tok/s131K ctx
dense
👁 Google
Gemma 2 9B
B66
9B15.4 GB41 tok/s8K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 14B
B66
14B16.3 GB33 tok/s69K ctx
dense
7B11.0 GB71 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 7B
B66
7B9.9 GB72 tok/s33K ctx
dense
👁 Cerebras
Cerebras-GPT 13B
B66
13B23.9 GB28 tok/s19K ctx
dense
👁 Baichuan
Baichuan 13B
B66
13B26.4 GB27 tok/s8K ctx
dense
👁 IBM
Granite Code 34B
B66
34B29.2 GB21 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 3B
B65
3B8.8 GB42 tok/s131K ctx
dense
👁 Cohere
Command R 35B
B65
35B28.6 GB21 tok/s4K ctx
dense
👁 TII
Falcon 7B Instruct
B65
7B9.2 GB72 tok/s8K ctx
dense
👁 Meta
Llama 3.2 11B Vision
B65
11B13.5 GB41 tok/s16K ctx
vision
👁 DeepSeek
DeepSeek R1 Distill 32B
B65
32B28.2 GB23 tok/s7K ctx
dense
👁 BigCode
StarCoder 15B
B64
15B30.2 GB20 tok/s8K ctx
dense
👁 Alibaba
Qwen 3 1.7B
B64
1.7B7.5 GB24 tok/s33K ctx
dense
👁 Mistral
Mistral Nemo 12B
B63
12B14.5 GB38 tok/s91K ctx
dense
👁 Mistral AI
Codestral 22B
B63
22B20.6 GB30 tok/s33K ctx
dense
👁 Microsoft
Phi 3 Medium 14B
B63
14B16.4 GB33 tok/s66K ctx
dense
👁 IBM
Granite Code 3B
B63
3B9.1 GB42 tok/s8K ctx
dense
👁 IBM
Granite 4.1 3B
B63
3B7.8 GB42 tok/s131K ctx
dense
👁 Mistral
Mistral 7B Instruct v0.3
B62
7B11.0 GB71 tok/s8K ctx
dense
👁 01.AI
Yi Coder 9B
B62
9B11.7 GB56 tok/s131K ctx
dense
B62
0.02B5.1 GB2 tok/s0K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 1.5B
B61
1.5B6.1 GB21 tok/s33K ctx
dense
👁 Mistral
Ministral 8B
B60
8B11.9 GB62 tok/s118K ctx
dense
👁 Meta
Llama 3.2 3B
B59
3B8.3 GB42 tok/s128K ctx
dense
👁 LMSYS
Vicuna 7B
B55
7B16.9 GB66 tok/s4K ctx
dense
👁 Cohere
Aya Expanse 32B
B55
32B26.7 GB25 tok/s8K ctx
dense
C

C Tier -- Tight Fit (187 models)

Runs with offloading or significant compromises.

👁 IBM
Granite 3.1 8B
C55
8B11.6 GB65 tok/s128K ctx
state-space
👁 01.AI
Yi 1.5 9B
C54
9B11.7 GB56 tok/s4K ctx
dense
C54
3B8.6 GB42 tok/s128K ctx
dense
👁 Alibaba
Qwen 2.5 Math 7B
C53
7B9.9 GB72 tok/s4K ctx
dense
👁 BigCode
StarCoder2 15B
C53
15B16.8 GB28 tok/s16K ctx
dense
Codestral 22B v0.1
C53
22B20.8 GB28 tok/s48K ctx
dense
Codestral 22B v0.1
C53
22B20.8 GB28 tok/s48K ctx
dense
👁 DeepSeek
DeepSeek R1 1.5B
C53
1.5B6.1 GB21 tok/s33K ctx
dense
Codestral 22B v0.1
C53
22B20.8 GB28 tok/s48K ctx
dense
👁 DeepSeek
DeepSeek LLM 7B
C53
7B16.4 GB66 tok/s4K ctx
dense
internlm2 5 20b chat
C53
20B19.3 GB28 tok/s61K ctx
dense
internlm2 limarp chat 20b
C53
20B19.3 GB28 tok/s61K ctx
dense
Codestral 22B v0.1 IMat
C53
22B20.8 GB28 tok/s48K ctx
dense
internlm2 math plus 20b i1
C53
20B19.3 GB28 tok/s61K ctx
dense
👁 TinyLlama
TinyLlama 1.1B
C53
1.1B5.8 GB15 tok/s4K ctx
dense
Codestral RAG 19B Pruned i1
C53
19B18.6 GB29 tok/s69K ctx
dense
Codestral 21B Pruned i1
C53
21B20.1 GB28 tok/s54K ctx
dense
Codestral 22B v0.1 i1
C52
22B20.8 GB28 tok/s48K ctx
dense
👁 OpenChat
OpenChat 7B
C52
7B11.0 GB71 tok/s8K ctx
dense
👁 Cohere
Aya Expanse 8B
C52
8B11.6 GB62 tok/s8K ctx
dense
👁 01.AI
Yi 1.5 34B
C52
34B29.2 GB21 tok/s4K ctx
dense
Phi 4 reasoning vision 15B
C51
15B15.7 GB30 tok/s109K ctx
dense
StarCoder2 15B
C51
15B15.7 GB30 tok/s109K ctx
dense
👁 Teknium
OpenHermes 2.5 7B
C51
7B11.0 GB71 tok/s8K ctx
dense
starcoder2 15b instruct v0.1
C51
15B15.7 GB30 tok/s109K ctx
dense
starcoder2 15b i1
C51
15B15.7 GB30 tok/s109K ctx
dense
8B11.6 GB62 tok/s33K ctx
dense
starcoder2 15b instruct v0.1
C51
15B15.7 GB30 tok/s109K ctx
dense
👁 Unsloth
DeepSeek R1 Distill Qwen 14B
C51
14B15.0 GB30 tok/s123K ctx
dense
HelpingAI 15B i1
C51
15B15.7 GB30 tok/s109K ctx
dense
👁 Hugging Face H4
Zephyr 7B Beta
C51
7B11.0 GB71 tok/s33K ctx
dense
NousResearch Hermes 4 14B
C51
14B15.0 GB30 tok/s123K ctx
dense
👁 UC Berkeley
Starling LM 7B
C50
7B11.0 GB71 tok/s8K ctx
dense
GGUF SOLARized GraniStral 14B 2102 YeAM HCT 32QKV
C50
14B15.0 GB30 tok/s123K ctx
dense
👁 Alibaba
Qwen 2.5 1.5B
C50
1.5B6.1 GB21 tok/s131K ctx
dense
internlm JanusCoder 14B
C50
14B15.0 GB30 tok/s123K ctx
dense
GGUF SOLARized GraniStral 14B 1902 YeAM HCT
C50
14B15.0 GB30 tok/s123K ctx
dense
👁 Unsloth
Qwen3.5 27B
C50
27B24.4 GB27 tok/s24K ctx
dense
👁 Unsloth
gemma 3 27b it
C50
27B24.4 GB27 tok/s24K ctx
dense
👁 Unsloth
Mistral Small 3.2 24B Instruct 2506
C50
24B22.2 GB27 tok/s37K ctx
dense
gemma 3 12b it
C50
12B13.5 GB35 tok/s157K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
C50
24B22.2 GB27 tok/s37K ctx
dense
Mistral Small 24B Instruct 2501
C50
24B22.2 GB27 tok/s37K ctx
dense
gemma 3 27b it
C50
27B24.4 GB27 tok/s24K ctx
dense
mistral small 3.1 24b instruct 2503 hf
C50
24B22.2 GB27 tok/s37K ctx
dense
cognitivecomputations Dolphin3.0 R1 Mistral 24B
C50
24B22.2 GB27 tok/s37K ctx
dense
Dolphin Mistral GLM 4.7 Flash 24B Venice Edition Thinking Uncensored i1
C50
24B22.2 GB27 tok/s37K ctx
dense
👁 Stability AI
StableLM 2 12B
C50
12B25.6 GB28 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 9B
C50
9B11.3 GB51 tok/s237K ctx
dense
Qwen3.5 9B Uncensored HauhauCS Aggressive
C50
9B11.3 GB51 tok/s237K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
C50
24B22.2 GB27 tok/s37K ctx
dense
Qwen3.5 9B
C50
9B11.3 GB51 tok/s237K ctx
dense
SOLAR 10.7B Instruct v1.0 uncensored
C50
10.7B12.6 GB40 tok/s186K ctx
dense
Nous Hermes 2 SOLAR 10.7B
C50
10.7B12.6 GB40 tok/s186K ctx
dense
Meta Llama 3.1 8B Instruct
C49
8B10.6 GB58 tok/s277K ctx
dense
llava llama 3 8b v1 1
C49
8B10.6 GB58 tok/s277K ctx
dense
👁 Intel
Neural Chat 7B
C49
7B11.0 GB71 tok/s8K ctx
dense
👁 Unsloth
DeepSeek R1 0528 Qwen3 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
Yi Coder 9B Chat
C49
9B11.3 GB51 tok/s237K ctx
dense
glm 4 9b chat 1m
C49
9B11.3 GB51 tok/s237K ctx
dense
Meta Llama 3 8B Instruct
C49
8B10.6 GB58 tok/s277K ctx
dense
👁 Unsloth
DeepSeek R1 Distill Llama 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
Dolphin3.0 Llama3.1 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
Llama 3 8B Instruct 32k v0.1
C49
8B10.6 GB58 tok/s277K ctx
dense
Llama 2 7B Chat
C49
7B9.9 GB66 tok/s329K ctx
dense
Meta Llama 3.1 8B Instruct
C49
8B10.6 GB58 tok/s277K ctx
dense
vntl llama3 8b v2
C49
8B10.6 GB58 tok/s277K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
Mistral 7B Instruct v0.2
C49
7B9.9 GB66 tok/s329K ctx
dense
Yi 1.5 9B Chat
C49
9B11.3 GB51 tok/s237K ctx
dense
solar finalised finetuned Model 10.7B i1
C49
10.7B12.6 GB40 tok/s186K ctx
dense
Mistral 7B Instruct v0.3
C49
7B9.9 GB66 tok/s329K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
Qwen3 8B DeepSeek v3.2 Speciale Distill
C49
8B10.6 GB58 tok/s277K ctx
dense
SOLAR 10.7B v1.0
C49
10.7B12.6 GB40 tok/s186K ctx
dense
👁 01.AI
Yi 1.5 6B
C49
6B9.4 GB84 tok/s4K ctx
dense
Hermes 3 Llama 3.1 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
HelpingAI2.5 10B i1
C49
10B12.1 GB46 tok/s205K ctx
dense
Hermes 2 Pro Llama 3 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
Dolphin3.0 Llama3.1 8B
C49
8B10.6 GB58 tok/s277K ctx
dense
granite 8b code instruct 4k
C49
8B10.6 GB58 tok/s277K ctx
dense
dolphin 2.9.4 llama3.1 8b
C49
8B10.6 GB58 tok/s277K ctx
dense
falcon mamba 7b instruct Q4 K M
C49
7B9.9 GB70 tok/s329K ctx
dense
👁 Google
Gemma 2 2B
C49
2B7.6 GB28 tok/s8K ctx
dense
Yi 1.5 6B Chat
C49
6B9.2 GB77 tok/s398K ctx
dense
Mistral 7B Instruct v0.3
C49
7B9.9 GB66 tok/s329K ctx
dense
Yi 9B Coder i1
C49
9B11.3 GB51 tok/s237K ctx
dense
Hermes 2 Pro Mistral 7B
C49
7B9.9 GB66 tok/s329K ctx
dense
openchat 3.6 8b 20240522 IMat
C49
8B10.6 GB58 tok/s277K ctx
dense
zephyr 7B beta
C49
7B9.9 GB66 tok/s329K ctx
dense
EXAONE 3.5 7.8B Instruct
C49
7.8B10.5 GB59 tok/s287K ctx
dense
aya expanse 8b
C49
8B10.6 GB58 tok/s277K ctx
dense
Nous Hermes 2 Mistral 7B DPO
C49
7B9.9 GB66 tok/s329K ctx
dense
HelpingAI2 9B
C49
9B11.3 GB51 tok/s237K ctx
dense
stablelm 2 zephyr 1 6b
C49
6B9.2 GB77 tok/s398K ctx
dense
EXAONE 3.5 7.8B Instruct
C49
7.8B10.5 GB59 tok/s287K ctx
dense
dolphin v2 8b abliterated i1
C49
8B10.6 GB58 tok/s277K ctx
dense
Falcon H1R 7B
C49
7B9.9 GB66 tok/s329K ctx
dense
EXAONE 3.5 7.8B Instruct i1
C49
7.8B10.5 GB59 tok/s287K ctx
dense
HelpingAI2 9B i1
C49
9B11.3 GB51 tok/s237K ctx
dense
exaone 3.0 7.8b it
C49
7.8B10.5 GB59 tok/s287K ctx
dense
Helply 10.2b chat i1
C49
10.2B12.2 GB42 tok/s200K ctx
dense
aya expanse 8b orthogonal heretic i1
C49
8B10.6 GB58 tok/s277K ctx
dense
HelpingAI 9B 200k i1
C49
9B11.3 GB51 tok/s237K ctx
dense
zephyr 7B alpha
C49
7B9.9 GB66 tok/s329K ctx
dense
speechless zephyr code functionary 7b
C49
7B9.9 GB66 tok/s329K ctx
dense
Mamba Codestral 7B v0.1
C49
7B9.9 GB70 tok/s329K ctx
dense
HelpingAI 9B i1
C49
9B11.3 GB51 tok/s237K ctx
dense
StarCoder2 7B
C49
7B9.9 GB66 tok/s329K ctx
dense
japanese stablelm instruct gamma 7B
C49
7B9.9 GB66 tok/s329K ctx
dense
aya expanse 8b orthogonal heretic
C49
8B10.6 GB58 tok/s277K ctx
dense
Yi 1.5 6B Chat
C49
6B9.2 GB77 tok/s398K ctx
dense
Falcon H1 7B Instruct
C49
7B9.9 GB66 tok/s329K ctx
dense
stablelm 2 1 6b chat imatrix
C49
6B9.2 GB77 tok/s398K ctx
dense
👁 Unsloth
Falcon H1R 7B
C49
7B9.9 GB66 tok/s329K ctx
dense
👁 NVIDIA
Nemotron Mini 4B
C49
4B9.2 GB56 tok/s4K ctx
dense
internlm2 5 1 8b chat i1
C49
8B10.6 GB58 tok/s277K ctx
dense
internlm3 8b instruct abliterated i1
C49
8B10.6 GB58 tok/s277K ctx
dense
👁 Google
Gemma 3 1B
C49
1B5.8 GB14 tok/s33K ctx
dense
stabilityai japanese stablelm base gamma 7b
C49
7B9.9 GB66 tok/s329K ctx
dense
baichuan2 7b chat
C49
7B9.9 GB66 tok/s329K ctx
dense
DiscoPOP zephyr 7b gemma
C49
7B9.9 GB66 tok/s329K ctx
dense
starcoder2 7b
C49
7B9.9 GB66 tok/s329K ctx
dense
zephyr 7b beta Mistral 7B Instruct v0.2
C49
7B9.9 GB66 tok/s329K ctx
dense
OpenChat 3.5 7B Qwen v2.0 i1
C49
7B9.9 GB66 tok/s329K ctx
dense
OpenChat 3.5 7B Starling v2.0 i1
C49
7B9.9 GB66 tok/s329K ctx
dense
internlm2 math plus 7b IMat
C49
7B9.9 GB66 tok/s329K ctx
dense
CodeNinja 1.0 OpenChat 7B i1
C49
7B9.9 GB66 tok/s329K ctx
dense
internlm2 5 7b chat i1
C49
7B9.9 GB66 tok/s329K ctx
dense
OpenSafetyLab MD Judge v0 2 internlm2 7b
C49
7B9.9 GB66 tok/s329K ctx
dense
MD Judge v0 2 internlm2 7b i1
C48
7B9.9 GB66 tok/s329K ctx
dense
zephyr 7b gemma sft african ultrachat 100k
C48
7B9.9 GB66 tok/s329K ctx
dense
jointpreferences mistral 7b sft helpful
C48
7B9.9 GB66 tok/s329K ctx
dense
HelpingAI2 6B
C48
6B9.2 GB77 tok/s398K ctx
dense
zephyr 7b dpo full i1
C48
7B9.9 GB66 tok/s329K ctx
dense
blossom v3 baichuan2 7b i1
C48
7B9.9 GB66 tok/s329K ctx
dense
blossom v1 baichuan 7b i1
C48
7B9.9 GB66 tok/s329K ctx
dense
HelpingAI2 6B i1
C48
6B9.2 GB77 tok/s398K ctx
dense
👁 BigCode
StarCoder2 7B
C48
7B9.5 GB72 tok/s16K ctx
dense
HelpingAI2.5 5B i1
C47
5B8.4 GB70 tok/s494K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 0.5B
C47
0.5B5.3 GB7 tok/s131K ctx
dense
👁 Unsloth
Qwen3.5 4B
C47
4B7.7 GB56 tok/s638K ctx
dense
gemma 3 4b it
C47
4B7.7 GB56 tok/s638K ctx
dense
gemma 3 4b it
C47
4B7.7 GB56 tok/s638K ctx
dense
Llama 3.2 3B Instruct
C46
3B7.3 GB42 tok/s863K ctx
dense
👁 Alibaba
Qwen 3 0.6B
C45
0.6B6.0 GB8 tok/s33K ctx
dense
👁 Qwen
Qwen2.5 3B Instruct
C45
3B7.0 GB42 tok/s878K ctx
dense
👁 Meta
Llama 3.2 1B
C45
1B5.9 GB14 tok/s128K ctx
dense
Llama 3.2 3B Instruct
C45
3B7.0 GB42 tok/s878K ctx
dense
Ministral 3 3B Instruct 2512
C45
3B7.0 GB42 tok/s878K ctx
dense
HELVETE 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
Hermes 3 Llama 3.2 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
stablelm zephyr 3b
C45
3B7.0 GB42 tok/s878K ctx
dense
StarCoder2 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
AI21 Jamba Reasoning 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
stablelm 3b 4e1t
C45
3B7.0 GB42 tok/s878K ctx
dense
ai21labs AI21 Jamba Reasoning 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
ai21labs AI21 Jamba2 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
HelpingAI 3B hindi i1
C45
3B7.0 GB42 tok/s878K ctx
dense
AI21 Jamba2 3B
C45
3B7.0 GB42 tok/s878K ctx
dense
HelpingAI 3B hindi
C45
3B7.0 GB42 tok/s878K ctx
dense
AI21 Jamba2 3B i1
C45
3B7.0 GB42 tok/s878K ctx
dense
gemma 2 2b it
C44
2B6.7 GB28 tok/s1331K ctx
dense
EXAONE 3.5 2.4B Instruct
C44
2.4B6.5 GB34 tok/s1119K ctx
dense
👁 BigCode
StarCoder2 3B
C44
3B7.1 GB42 tok/s16K ctx
dense
👁 Google
gemma 2b
C44
2B6.2 GB28 tok/s1359K ctx
dense
Gemmasutra Mini 2B v1
C43
2B6.2 GB28 tok/s1359K ctx
dense
gemma 2 2b it
C43
2B6.2 GB28 tok/s1359K ctx
dense
👁 Alibaba
Qwen 3.5 0.6B
C43
0.6B6.0 GB8 tok/s131K ctx
dense
👁 Qwen
Qwen2.5 1.5B Instruct
C42
1.5B5.9 GB21 tok/s1840K ctx
dense
👁 Unsloth
DeepSeek R1 Distill Qwen 1.5B
C42
1.5B5.9 GB21 tok/s1840K ctx
dense
Yi Coder 1.5B Chat
C42
1.5B5.9 GB21 tok/s1840K ctx
dense
stablelm 2 zephyr 1.6b
C42
1.6B6.0 GB22 tok/s1720K ctx
dense
logos16v2 stablelm2 1.6b i1
C42
1.6B6.0 GB22 tok/s1720K ctx
dense
Yi Coder 1.5B
C42
1.5B5.9 GB21 tok/s1840K ctx
dense
👁 Unsloth
Falcon H1 1.5B Instruct
C42
1.5B5.9 GB21 tok/s1840K ctx
dense
TinyLlama 1.1B Chat v1.0
C41
1.1B5.6 GB15 tok/s2540K ctx
dense
EXAONE 4.0 1.2B
C41
1.2B5.7 GB17 tok/s2321K ctx
dense
Llama 3.2 1B Instruct Q8 0
C41
1B5.7 GB14 tok/s2773K ctx
dense
👁 Alibaba
Qwen 2.5 0.5B
C41
0.5B5.3 GB7 tok/s131K ctx
dense
TinyLlama 1.1B Chat v0.3
C41
1.1B5.6 GB15 tok/s2540K ctx
dense
👁 TinyLlama
TinyLlama 1.1B Chat v0.6
C41
1.1B5.6 GB15 tok/s2540K ctx
dense
👁 01.AI
Yi 34B Chat
C41
34B29.2 GB21 tok/s4K ctx
dense
Llama 3.2 1B Instruct
C41
1B5.5 GB14 tok/s2802K ctx
dense
gemma 3 1b it
C40
1B5.5 GB14 tok/s2802K ctx
dense
TinyLlama 1.1B Chat v1.0 imatrix
C40
1.1B5.6 GB15 tok/s2540K ctx
dense
Falcon3 1B Instruct abliterated
C40
1B5.5 GB14 tok/s2802K ctx
dense
F

F Tier -- Can't Run (69 models)

Too heavy for this hardware.

👁 Unsloth
Qwen3.5 35B A3B
D39
35B30.2 GB18 tok/s4K ctx
dense
EXAONE 4.0 32B
D39
32B28.1 GB21 tok/s7K ctx
dense
aya expanse 32b heretic MPOA i1
D39
32B28.1 GB21 tok/s7K ctx
dense
Baichuan M2 32B Q4 K M
D39
32B28.1 GB21 tok/s7K ctx
dense
Qwen3.5 35B A3B
D39
35B30.2 GB18 tok/s4K ctx
dense
baichuan inc Baichuan M2 32B
D39
32B28.1 GB21 tok/s7K ctx
dense
SmolVLM 500M Instruct
D38
0.5B5.3 GB7 tok/s3316K ctx
dense
embeddinggemma 300M
D37
0.3B5.1 GB4 tok/s3342K ctx
dense
granite embedding 107m multilingual
D35
0.11B5.0 GB2 tok/s3371K ctx
dense
Falcon H1 Tiny 90M Instruct
D34
0.09B4.9 GB2 tok/s3372K ctx
dense
Hermes 4.3 36B
D32
36B31.0 GB17 tok/s4K ctx
dense
👁 Meta
Llama 3.1 70B
F0
70B52.4 GB6 tok/s4K ctx
dense
👁 Meta
Llama 3.3 70B
F0
70B52.4 GB6 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 72B
F0
72B53.6 GB5 tok/s4K ctx
dense
👁 Alibaba
Qwen 3 235B A22B
F0
235B151.0 GB2 tok/s4K ctx
moe
👁 Alibaba
Qwen3-Coder-Next
F0
80B55.1 GB9 tok/s4K ctx
moe
👁 Alibaba
Qwen3-Coder 480B A35B Instruct
F0
480B300.5 GB2 tok/s4K ctx
moe
👁 Alibaba
Qwen 3.5 122B A10B
F0
122B81.6 GB7 tok/s4K ctx
moe
👁 Alibaba
Qwen 3.5 397B A17B
F0
397B249.8 GB2 tok/s4K ctx
moe
👁 Google
Gemma 2 27B
F0
27B32.5 GB15 tok/s7K ctx
dense
👁 Google
Gemma 3 27B
F0
27B32.5 GB15 tok/s7K ctx
dense
👁 Mistral
Mixtral 8x7B
F0
47B35.4 GB12 tok/s4K ctx
moe
👁 Mistral
Mixtral 8x22B
F0
141B94.2 GB3 tok/s4K ctx
moe
👁 TII
Falcon 40B Instruct
F0
40B35.4 GB12 tok/s4K ctx
dense
👁 MosaicML
MPT-30B-Instruct
F0
30B49.8 GB10 tok/s4K ctx
dense
👁 Mistral
Devstral 2 123B Instruct
F0
123B85.2 GB3 tok/s4K ctx
dense
👁 Mistral
Mistral Small 4 119B
F0
119B82.7 GB8 tok/s4K ctx
moe
👁 Mistral
Leanstral 119B A6B
F0
119B86.2 GB7 tok/s4K ctx
moe
👁 Mistral
Mistral Large 3
F0
675B421.9 GB2 tok/s4K ctx
+1moe
F0
744B477.7 GB2 tok/s4K ctx
moe
👁 Z.ai
GLM-5.1
F0
754B483.8 GB2 tok/s4K ctx
moe
F0
1000B622.2 GB2 tok/s4K ctx
moe
F0
1000B622.2 GB2 tok/s4K ctx
+1moe
👁 Moonshot AI
Kimi Linear 48B A3B
F0
48B35.0 GB11 tok/s4K ctx
linear
👁 InternLM
InternLM 20B
F0
20B37.5 GB15 tok/s6K ctx
dense
👁 DeepSeek
DeepSeek V3 671B
F0
671B473.7 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V3.1 671B
F0
671B473.7 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V4 Pro
F0
1600B868.6 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V4 Flash
F0
284B164.1 GB4 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek R1 671B
F0
671B473.7 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek Coder V2 236B
F0
236B207.3 GB3 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V2.5 236B
F0
236B207.3 GB3 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek LLM 67B
F0
67B51.5 GB6 tok/s4K ctx
dense
👁 Meta
Llama 4 Scout 17B 16E
F0
109B74.2 GB4 tok/s4K ctx
moe
👁 Meta
Llama 4 Maverick 17B 128E
F0
400B251.7 GB2 tok/s4K ctx
moe
👁 Cohere
Command R+ 104B
F0
104B71.6 GB4 tok/s4K ctx
dense
👁 Cohere
Command A 111B
F0
111B76.4 GB4 tok/s4K ctx
dense
👁 Mistral AI
Pixtral Large 124B
F0
124B85.8 GB3 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 VL 72B
F0
72B53.6 GB5 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 Math 72B
F0
72B53.6 GB5 tok/s4K ctx
dense
👁 NVIDIA
Nemotron 70B
F0
70B52.4 GB6 tok/s4K ctx
dense
👁 Meta
Llama 3.1 405B
F0
405B259.5 GB2 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 70B
F0
70B52.4 GB6 tok/s4K ctx
dense
👁 OpenAI
GPT-OSS 120B
F0
117B81.0 GB3 tok/s4K ctx
dense
👁 Google
Gemma 4 31B
F0
30.7B38.2 GB12 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek V3.2
F0
671B414.6 GB2 tok/s4K ctx
moe
MiniMax M2.7
F0
230B148.9 GB4 tok/s4K ctx
moe
Solar Open 100B
F0
100B77.5 GB4 tok/s4K ctx
dense
Qwen3 48B A4B Savant Commander Distill 12X Closed Open Heretic Uncensored
F0
48B39.7 GB9 tok/s4K ctx
dense
K EXAONE 236B A23B
F0
236B176.4 GB2 tok/s4K ctx
dense
Llama 3.3 70B Instruct
F0
70B55.7 GB5 tok/s4K ctx
dense
Baichuan M3 235B
F0
235B175.7 GB2 tok/s4K ctx
dense
Baichuan M3 235B i1
F0
235B175.7 GB2 tok/s4K ctx
dense
BaichuanMed OCR 72B i1
F0
72B57.1 GB5 tok/s4K ctx
dense
Solar Open 100B i1
F0
100B77.5 GB4 tok/s4K ctx
dense
Solar Open 69B REAP i1
F0
69B55.0 GB5 tok/s4K ctx
dense
stabilityai japanese stablelm instruct beta 70b
F0
70B55.7 GB5 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 122B A10B
F0
122B78.9 GB3 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 397B A17B
F0
397B293.5 GB2 tok/s4K ctx
dense

Compare with...

See How Other Hardware Stacks Up

Check tier lists for similar GPUs and Apple Silicon configs

MacBook Pro M3 Pro 36GB

36 GB unified memory

View tier list

MacBook Pro M1 Max 32GB

32 GB unified memory

View tier list

MacBook Pro M1 Pro 32GB

32 GB unified memory

View tier list

MacBook Pro M2 Pro 32GB

32 GB unified memory

View tier list
Full MacBook Pro M4 Max 36GB pageCompare this hardware