VOOZH about

URL: https://willitrunai.com/my-results/m4-pro-24gb

⇱ MacBook Pro M4 Pro 24GB AI Tier List: 257 models ranked | WillItRunAI


AI Compatibility Tier List

MacBook Pro M4 Pro 24GB

374 AI models ranked by how well they run on your 24 GB unified memory hardware. Best match: Qwen 3.5 9B (score: 94).

374

Models tested

257

Can run

24 GB

unified memory

38

Best tok/s

Tier Distribution

S 7
A 50
B 33
C 167
F 117
S: 7A: 50B: 33C: 167F: 117
S

S Tier -- Perfect Fit (7 models)

These models run flawlessly with headroom to spare.

👁 Alibaba
Qwen 3.5 9B
S94
9B11.2 GB38 tok/s60K ctx
dense
👁 Alibaba
Qwen 3 8B
S92
8B10.6 GB43 tok/s65K ctx
dense
👁 Alibaba
Qwen 3.5 4B
S90
4B8.1 GB56 tok/s83K ctx
dense
👁 Alibaba
Qwen 3 14B
S90
14B14.5 GB23 tok/s34K ctx
dense
👁 Microsoft
Phi-4-reasoning-plus 14B
S89
14.7B15.5 GB23 tok/s25K ctx
dense
👁 NVIDIA
Nemotron Nano 8B
S87
8B10.3 GB46 tok/s73K ctx
dense
👁 Microsoft
Phi-4 Mini Reasoning 4B
S86
3.8B7.3 GB53 tok/s125K ctx
dense
A

A Tier -- Great Fit (50 models)

Excellent performance with comfortable memory margins.

👁 Mistral
Ministral 3 14B
A84
14B14.5 GB23 tok/s34K ctx
multimodal
👁 InternLM
InternVL2 8B
A84
8B10.3 GB43 tok/s8K ctx
dense
👁 Mistral
Ministral 3 8B
A83
8B10.6 GB43 tok/s65K ctx
multimodal
👁 NVIDIA
Nemotron Nano 9B v2
A83
9B11.4 GB41 tok/s54K ctx
dense
A83
19B17.5 GB24 tok/s8K ctx
dense
👁 OpenBMB
MiniCPM-V 2.6 8B
A83
8B10.3 GB43 tok/s2K ctx
dense
👁 Alibaba
Qwen 3 4B
A82
4B8.1 GB56 tok/s33K ctx
dense
👁 Defog
SQLCoder 7B
A81
7B9.7 GB49 tok/s8K ctx
dense
A81
14B15.1 GB23 tok/s16K ctx
dense
👁 DeepSeek
DeepSeek Coder V2 16B
A81
16B16.5 GB51 tok/s20K ctx
moe
👁 Alibaba
Qwen 2.5 14B
A80
14B15.0 GB23 tok/s29K ctx
dense
👁 Alibaba
Qwen 2.5 VL 7B
A80
7B8.6 GB49 tok/s33K ctx
dense
👁 Magistral AI
Magistral 7B
A80
7B9.7 GB49 tok/s8K ctx
dense
👁 AllenAI
OLMo 2 13B
A80
13B13.9 GB25 tok/s33K ctx
dense
👁 OpenAI
GPT-OSS 20B
A79
21B18.7 GB35 tok/s6K ctx
moe
👁 Jina AI
Jina Embeddings v3
A79
0.57B6.6 GB8 tok/s8K ctx
dense
👁 Tsinghua/Zhipu
CodeGeeX 4 9B
A79
9B9.6 GB39 tok/s131K ctx
dense
👁 Google
Gemma 3 12B
A79
12B15.7 GB20 tok/s21K ctx
dense
👁 Google
Gemma 4 E4B
A78
8B9.7 GB32 tok/s111K ctx
dense
👁 IBM
Granite Code 8B
A77
8B10.3 GB43 tok/s8K ctx
dense
👁 BAAI
BGE M3
A77
0.57B5.8 GB8 tok/s8K ctx
dense
👁 IBM
Granite 4.1 8B
A77
8B10.8 GB43 tok/s58K ctx
dense
👁 DevStral AI
DevStral 7B
A77
7B9.7 GB49 tok/s8K ctx
dense
👁 Mistral AI
Pixtral 12B
A77
12B13.3 GB27 tok/s42K ctx
dense
👁 Alibaba
Qwen 2.5 7B
A76
7B8.6 GB49 tok/s131K ctx
dense
👁 Mistral AI
Codestral Mamba 7B
A76
7B8.3 GB57 tok/s262K ctx
state-space
👁 BigCode
StarCoder 7B
A75
7B15.1 GB45 tok/s8K ctx
dense
👁 Meta
CodeLlama 7B Instruct
A75
7B15.6 GB45 tok/s16K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 3B
A74
3B7.5 GB42 tok/s87K ctx
dense
👁 Mistral
Magistral Small 2507
A74
24B20.6 GB18 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 14B
A74
14B15.0 GB23 tok/s29K ctx
dense
👁 Mistral
Devstral Small 2 24B Instruct
A74
24B20.6 GB18 tok/s4K ctx
dense
A74
7B10.7 GB49 tok/s8K ctx
dense
👁 Mixedbread AI
mxbai Embed Large
A74
0.34B5.6 GB5 tok/s1K ctx
dense
👁 Meta
Llama 3.1 8B
A73
8B10.3 GB43 tok/s73K ctx
dense
👁 Mistral
Codestral 2 25.08
A73
22B19.4 GB18 tok/s4K ctx
dense
👁 Snowflake
Snowflake Arctic Embed L
A73
0.34B5.6 GB5 tok/s1K ctx
dense
👁 WizardLM
WizardMath 7B
A72
7B9.7 GB49 tok/s4K ctx
dense
👁 Allen AI
OLMo 2 7B
A72
7B9.7 GB49 tok/s4K ctx
dense
👁 Zhipu
GLM-4 9B
A72
9B9.6 GB39 tok/s128K ctx
dense
👁 InternLM
InternLM 7B
A72
7B15.6 GB45 tok/s8K ctx
dense
👁 InternLM
InternLM Chat 7B
A72
7B15.6 GB45 tok/s8K ctx
dense
👁 Mistral
Devstral Small 1.1
A72
24B20.6 GB18 tok/s4K ctx
dense
👁 Google
Gemma 4 E2B
A72
5.1B7.1 GB51 tok/s128K ctx
dense
👁 Microsoft
Phi 3 Mini 3.8B
A72
3.8B11.7 GB53 tok/s31K ctx
dense
👁 Google
Gemma 3 4B
A71
4B8.0 GB56 tok/s87K ctx
dense
👁 BAAI
BGE Large EN v1.5
A71
0.34B5.6 GB5 tok/s1K ctx
dense
👁 Microsoft
Phi 3.5 Mini 4B
A71
4B11.8 GB56 tok/s31K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 7B
A70
7B8.6 GB49 tok/s131K ctx
dense
👁 Microsoft
Phi 4 Mini 4B
A70
4B7.4 GB56 tok/s124K ctx
dense
B

B Tier -- Good Fit (33 models)

Solid performance, may need quantization or reduced context.

👁 LLaVA
LLaVA 1.5 7B
B70
7B15.6 GB45 tok/s4K ctx
dense
👁 Mistral
Ministral 3 3B
B69
3B6.1 GB42 tok/s261K ctx
multimodal
👁 IBM
Granite Code 20B
B69
20B18.9 GB21 tok/s8K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 8B
B69
8B10.3 GB43 tok/s33K ctx
dense
👁 Alibaba
Qwen 3.5 2B
B68
2B6.4 GB28 tok/s118K ctx
dense
👁 Nomic AI
Nomic Embed Text v1.5
B68
0.14B4.3 GB2 tok/s8K ctx
dense
7B9.7 GB49 tok/s4K ctx
dense
👁 MosaicML
MPT-7B-Instruct
B68
7B15.6 GB45 tok/s8K ctx
dense
👁 Baichuan
Baichuan 7B
B68
7B15.6 GB45 tok/s8K ctx
dense
👁 Meta
Llama 3.2 11B Vision
B68
11B12.2 GB30 tok/s16K ctx
vision
👁 Google
Gemma 2 9B
B68
9B14.1 GB28 tok/s8K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 7B
B68
7B8.6 GB49 tok/s33K ctx
dense
👁 Alibaba
Qwen 2.5 3B
B67
3B7.5 GB42 tok/s87K ctx
dense
👁 Mistral
Mistral Small 3.2 24B
B67
24B20.6 GB18 tok/s4K ctx
vision
B66
24B20.6 GB18 tok/s4K ctx
dense
👁 TII
Falcon 7B Instruct
B66
7B7.9 GB50 tok/s8K ctx
dense
👁 Mistral
Mistral Nemo 12B
B65
12B13.3 GB27 tok/s42K ctx
dense
👁 IBM
Granite Code 3B
B65
3B7.8 GB42 tok/s8K ctx
dense
👁 Alibaba
Qwen 3 1.7B
B65
1.7B6.2 GB24 tok/s33K ctx
dense
👁 Mistral
Mistral Small 24B
B65
24B20.6 GB18 tok/s4K ctx
dense
👁 IBM
Granite 4.1 3B
B64
3B6.5 GB42 tok/s131K ctx
dense
👁 01.AI
Yi Coder 9B
B64
9B10.4 GB38 tok/s91K ctx
dense
👁 Mistral
Mistral 7B Instruct v0.3
B64
7B9.7 GB49 tok/s8K ctx
dense
👁 Mistral AI
Mistral Small 3.1 24B
B64
24B20.6 GB18 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 14B
B64
14B15.0 GB23 tok/s29K ctx
dense
B62
0.02B3.8 GB2 tok/s0K ctx
dense
👁 Mistral
Ministral 8B
B62
8B10.6 GB43 tok/s65K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 1.5B
B62
1.5B4.8 GB21 tok/s33K ctx
dense
👁 Meta
Llama 3.2 3B
B61
3B7.0 GB42 tok/s112K ctx
dense
👁 Microsoft
Phi 3 Medium 14B
B60
14B15.1 GB23 tok/s28K ctx
dense
👁 IBM
Granite 3.1 8B
B57
8B10.3 GB53 tok/s73K ctx
state-space
👁 01.AI
Yi 1.5 9B
B56
9B10.4 GB38 tok/s4K ctx
dense
B56
3B7.3 GB42 tok/s98K ctx
dense
C

C Tier -- Tight Fit (167 models)

Runs with offloading or significant compromises.

👁 Alibaba
Qwen 2.5 Math 7B
C55
7B8.6 GB49 tok/s4K ctx
dense
👁 OpenChat
OpenChat 7B
C54
7B9.7 GB49 tok/s8K ctx
dense
👁 DeepSeek
DeepSeek R1 1.5B
C54
1.5B4.8 GB21 tok/s33K ctx
dense
👁 Cohere
Aya Expanse 8B
C54
8B10.3 GB43 tok/s8K ctx
dense
👁 TinyLlama
TinyLlama 1.1B
C53
1.1B4.5 GB15 tok/s4K ctx
dense
gemma 3 12b it
C53
12B12.2 GB25 tok/s74K ctx
dense
8B10.3 GB43 tok/s33K ctx
dense
👁 Teknium
OpenHermes 2.5 7B
C53
7B9.7 GB49 tok/s8K ctx
dense
👁 Hugging Face H4
Zephyr 7B Beta
C52
7B9.7 GB49 tok/s33K ctx
dense
👁 Unsloth
DeepSeek R1 Distill Qwen 14B
C52
14B13.7 GB22 tok/s51K ctx
dense
👁 LMSYS
Vicuna 7B
C52
7B15.6 GB45 tok/s4K ctx
dense
👁 UC Berkeley
Starling LM 7B
C52
7B9.7 GB49 tok/s8K ctx
dense
NousResearch Hermes 4 14B
C52
14B13.7 GB22 tok/s51K ctx
dense
SOLAR 10.7B Instruct v1.0 uncensored
C52
10.7B11.3 GB28 tok/s93K ctx
dense
Nous Hermes 2 SOLAR 10.7B
C52
10.7B11.3 GB28 tok/s93K ctx
dense
GGUF SOLARized GraniStral 14B 2102 YeAM HCT 32QKV
C52
14B13.7 GB22 tok/s51K ctx
dense
internlm JanusCoder 14B
C52
14B13.7 GB22 tok/s51K ctx
dense
GGUF SOLARized GraniStral 14B 1902 YeAM HCT
C52
14B13.7 GB22 tok/s51K ctx
dense
solar finalised finetuned Model 10.7B i1
C52
10.7B11.3 GB28 tok/s93K ctx
dense
SOLAR 10.7B v1.0
C52
10.7B11.3 GB28 tok/s93K ctx
dense
👁 Unsloth
Qwen3.5 9B
C52
9B10.0 GB35 tok/s126K ctx
dense
Qwen3.5 9B Uncensored HauhauCS Aggressive
C52
9B10.0 GB35 tok/s126K ctx
dense
Qwen3.5 9B
C51
9B10.0 GB35 tok/s126K ctx
dense
HelpingAI2.5 10B i1
C51
10B10.8 GB32 tok/s105K ctx
dense
👁 Alibaba
Qwen 2.5 1.5B
C51
1.5B4.8 GB21 tok/s131K ctx
dense
Yi Coder 9B Chat
C51
9B10.0 GB35 tok/s126K ctx
dense
glm 4 9b chat 1m
C51
9B10.0 GB35 tok/s126K ctx
dense
Helply 10.2b chat i1
C51
10.2B10.9 GB30 tok/s101K ctx
dense
👁 Intel
Neural Chat 7B
C51
7B9.7 GB49 tok/s8K ctx
dense
Yi 1.5 9B Chat
C51
9B10.0 GB35 tok/s126K ctx
dense
Meta Llama 3.1 8B Instruct
C51
8B9.3 GB40 tok/s152K ctx
dense
llava llama 3 8b v1 1
C51
8B9.3 GB40 tok/s152K ctx
dense
👁 Unsloth
DeepSeek R1 0528 Qwen3 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
Meta Llama 3 8B Instruct
C51
8B9.3 GB40 tok/s152K ctx
dense
👁 Unsloth
DeepSeek R1 Distill Llama 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
Dolphin3.0 Llama3.1 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
👁 NVIDIA
Nemotron Mini 4B
C51
4B7.9 GB56 tok/s4K ctx
dense
Llama 3 8B Instruct 32k v0.1
C51
8B9.3 GB40 tok/s152K ctx
dense
Meta Llama 3.1 8B Instruct
C51
8B9.3 GB40 tok/s152K ctx
dense
vntl llama3 8b v2
C51
8B9.3 GB40 tok/s152K ctx
dense
falcon mamba 7b instruct Q4 K M
C51
7B8.6 GB57 tok/s186K ctx
dense
Yi 9B Coder i1
C51
9B10.0 GB35 tok/s126K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
HelpingAI2 9B
C51
9B10.0 GB35 tok/s126K ctx
dense
Qwen3 8B DeepSeek v3.2 Speciale Distill
C51
8B9.3 GB40 tok/s152K ctx
dense
Hermes 3 Llama 3.1 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
HelpingAI2 9B i1
C51
9B10.0 GB35 tok/s126K ctx
dense
HelpingAI 9B 200k i1
C51
9B10.0 GB35 tok/s126K ctx
dense
HelpingAI 9B i1
C51
9B10.0 GB35 tok/s126K ctx
dense
Hermes 2 Pro Llama 3 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
Dolphin3.0 Llama3.1 8B
C51
8B9.3 GB40 tok/s152K ctx
dense
Mamba Codestral 7B v0.1
C51
7B8.6 GB57 tok/s186K ctx
dense
granite 8b code instruct 4k
C51
8B9.3 GB40 tok/s152K ctx
dense
dolphin 2.9.4 llama3.1 8b
C51
8B9.3 GB40 tok/s152K ctx
dense
Llama 2 7B Chat
C51
7B8.6 GB45 tok/s186K ctx
dense
👁 Google
Gemma 2 2B
C51
2B6.3 GB28 tok/s8K ctx
dense
Mistral 7B Instruct v0.2
C51
7B8.6 GB45 tok/s186K ctx
dense
openchat 3.6 8b 20240522 IMat
C50
8B9.3 GB40 tok/s152K ctx
dense
Mistral 7B Instruct v0.3
C50
7B8.6 GB45 tok/s186K ctx
dense
aya expanse 8b
C50
8B9.3 GB40 tok/s152K ctx
dense
EXAONE 3.5 7.8B Instruct
C50
7.8B9.2 GB41 tok/s158K ctx
dense
dolphin v2 8b abliterated i1
C50
8B9.3 GB40 tok/s152K ctx
dense
EXAONE 3.5 7.8B Instruct
C50
7.8B9.2 GB41 tok/s158K ctx
dense
aya expanse 8b orthogonal heretic i1
C50
8B9.3 GB40 tok/s152K ctx
dense
EXAONE 3.5 7.8B Instruct i1
C50
7.8B9.2 GB41 tok/s158K ctx
dense
👁 BigCode
StarCoder2 15B
C50
15B15.5 GB20 tok/s16K ctx
dense
exaone 3.0 7.8b it
C50
7.8B9.2 GB41 tok/s158K ctx
dense
aya expanse 8b orthogonal heretic
C50
8B9.3 GB40 tok/s152K ctx
dense
Mistral 7B Instruct v0.3
C50
7B8.6 GB45 tok/s186K ctx
dense
Hermes 2 Pro Mistral 7B
C50
7B8.6 GB45 tok/s186K ctx
dense
👁 01.AI
Yi 1.5 6B
C50
6B8.1 GB58 tok/s4K ctx
dense
zephyr 7B beta
C50
7B8.6 GB45 tok/s186K ctx
dense
internlm2 5 1 8b chat i1
C50
8B9.3 GB40 tok/s152K ctx
dense
internlm3 8b instruct abliterated i1
C50
8B9.3 GB40 tok/s152K ctx
dense
Nous Hermes 2 Mistral 7B DPO
C50
7B8.6 GB45 tok/s186K ctx
dense
Falcon H1R 7B
C50
7B8.6 GB45 tok/s186K ctx
dense
zephyr 7B alpha
C50
7B8.6 GB45 tok/s186K ctx
dense
speechless zephyr code functionary 7b
C50
7B8.6 GB45 tok/s186K ctx
dense
StarCoder2 7B
C50
7B8.6 GB45 tok/s186K ctx
dense
japanese stablelm instruct gamma 7B
C50
7B8.6 GB45 tok/s186K ctx
dense
Falcon H1 7B Instruct
C50
7B8.6 GB45 tok/s186K ctx
dense
👁 Unsloth
Falcon H1R 7B
C50
7B8.6 GB45 tok/s186K ctx
dense
👁 DeepSeek
DeepSeek LLM 7B
C50
7B15.1 GB45 tok/s4K ctx
dense
Yi 1.5 6B Chat
C50
6B7.9 GB53 tok/s230K ctx
dense
stabilityai japanese stablelm base gamma 7b
C50
7B8.6 GB45 tok/s186K ctx
dense
stablelm 2 zephyr 1 6b
C50
6B7.9 GB53 tok/s230K ctx
dense
baichuan2 7b chat
C50
7B8.6 GB45 tok/s186K ctx
dense
DiscoPOP zephyr 7b gemma
C50
7B8.6 GB45 tok/s186K ctx
dense
starcoder2 7b
C50
7B8.6 GB45 tok/s186K ctx
dense
zephyr 7b beta Mistral 7B Instruct v0.2
C50
7B8.6 GB45 tok/s186K ctx
dense
OpenChat 3.5 7B Qwen v2.0 i1
C50
7B8.6 GB45 tok/s186K ctx
dense
OpenChat 3.5 7B Starling v2.0 i1
C50
7B8.6 GB45 tok/s186K ctx
dense
internlm2 math plus 7b IMat
C50
7B8.6 GB45 tok/s186K ctx
dense
CodeNinja 1.0 OpenChat 7B i1
C50
7B8.6 GB45 tok/s186K ctx
dense
internlm2 5 7b chat i1
C50
7B8.6 GB45 tok/s186K ctx
dense
OpenSafetyLab MD Judge v0 2 internlm2 7b
C50
7B8.6 GB45 tok/s186K ctx
dense
MD Judge v0 2 internlm2 7b i1
C50
7B8.6 GB45 tok/s186K ctx
dense
zephyr 7b gemma sft african ultrachat 100k
C50
7B8.6 GB45 tok/s186K ctx
dense
jointpreferences mistral 7b sft helpful
C50
7B8.6 GB45 tok/s186K ctx
dense
zephyr 7b dpo full i1
C50
7B8.6 GB45 tok/s186K ctx
dense
Yi 1.5 6B Chat
C50
6B7.9 GB53 tok/s230K ctx
dense
blossom v3 baichuan2 7b i1
C50
7B8.6 GB45 tok/s186K ctx
dense
stablelm 2 1 6b chat imatrix
C50
6B7.9 GB53 tok/s230K ctx
dense
blossom v1 baichuan 7b i1
C50
7B8.6 GB45 tok/s186K ctx
dense
HelpingAI2 6B
C49
6B7.9 GB53 tok/s230K ctx
dense
👁 Google
Gemma 3 1B
C49
1B4.5 GB14 tok/s33K ctx
dense
HelpingAI2 6B i1
C49
6B7.9 GB53 tok/s230K ctx
dense
👁 BigCode
StarCoder2 7B
C49
7B8.3 GB49 tok/s16K ctx
dense
Phi 4 reasoning vision 15B
C49
15B14.4 GB21 tok/s42K ctx
dense
StarCoder2 15B
C49
15B14.4 GB21 tok/s42K ctx
dense
HelpingAI2.5 5B i1
C49
5B7.1 GB63 tok/s293K ctx
dense
👁 Mistral AI
Codestral 22B
C49
22B19.4 GB20 tok/s4K ctx
dense
starcoder2 15b instruct v0.1
C49
15B14.4 GB21 tok/s42K ctx
dense
starcoder2 15b i1
C49
15B14.4 GB21 tok/s42K ctx
dense
Codestral RAG 19B Pruned i1
C49
19B17.3 GB23 tok/s16K ctx
dense
starcoder2 15b instruct v0.1
C49
15B14.4 GB21 tok/s42K ctx
dense
HelpingAI 15B i1
C49
15B14.4 GB21 tok/s42K ctx
dense
internlm2 5 20b chat
C49
20B18.0 GB21 tok/s11K ctx
dense
internlm2 limarp chat 20b
C49
20B18.0 GB21 tok/s11K ctx
dense
internlm2 math plus 20b i1
C49
20B18.0 GB21 tok/s11K ctx
dense
👁 Unsloth
Qwen3.5 4B
C49
4B6.4 GB56 tok/s387K ctx
dense
gemma 3 4b it
C48
4B6.4 GB56 tok/s387K ctx
dense
gemma 3 4b it
C48
4B6.4 GB56 tok/s387K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 0.5B
C48
0.5B4.0 GB7 tok/s131K ctx
dense
Llama 3.2 3B Instruct
C47
3B6.0 GB42 tok/s529K ctx
dense
👁 Qwen
Qwen2.5 3B Instruct
C47
3B5.7 GB42 tok/s544K ctx
dense
Llama 3.2 3B Instruct
C47
3B5.7 GB42 tok/s544K ctx
dense
Ministral 3 3B Instruct 2512
C46
3B5.7 GB42 tok/s544K ctx
dense
👁 Alibaba
Qwen 3 0.6B
C46
0.6B4.7 GB8 tok/s33K ctx
dense
HELVETE 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
Hermes 3 Llama 3.2 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
stablelm zephyr 3b
C46
3B5.7 GB42 tok/s544K ctx
dense
StarCoder2 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
👁 Meta
Llama 3.2 1B
C46
1B4.6 GB14 tok/s128K ctx
dense
AI21 Jamba Reasoning 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
stablelm 3b 4e1t
C46
3B5.7 GB42 tok/s544K ctx
dense
ai21labs AI21 Jamba Reasoning 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
ai21labs AI21 Jamba2 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
HelpingAI 3B hindi i1
C46
3B5.7 GB42 tok/s544K ctx
dense
AI21 Jamba2 3B
C46
3B5.7 GB42 tok/s544K ctx
dense
HelpingAI 3B hindi
C46
3B5.7 GB42 tok/s544K ctx
dense
AI21 Jamba2 3B i1
C46
3B5.7 GB42 tok/s544K ctx
dense
gemma 2 2b it
C45
2B5.4 GB28 tok/s829K ctx
dense
👁 BigCode
StarCoder2 3B
C45
3B5.8 GB42 tok/s16K ctx
dense
EXAONE 3.5 2.4B Instruct
C45
2.4B5.2 GB34 tok/s701K ctx
dense
👁 Google
gemma 2b
C44
2B4.9 GB28 tok/s858K ctx
dense
Gemmasutra Mini 2B v1
C44
2B4.9 GB28 tok/s858K ctx
dense
gemma 2 2b it
C44
2B4.9 GB28 tok/s858K ctx
dense
👁 Alibaba
Qwen 3.5 0.6B
C44
0.6B4.7 GB8 tok/s131K ctx
dense
👁 Qwen
Qwen2.5 1.5B Instruct
C43
1.5B4.6 GB21 tok/s1172K ctx
dense
👁 Unsloth
DeepSeek R1 Distill Qwen 1.5B
C43
1.5B4.6 GB21 tok/s1172K ctx
dense
Yi Coder 1.5B Chat
C43
1.5B4.6 GB21 tok/s1172K ctx
dense
stablelm 2 zephyr 1.6b
C43
1.6B4.7 GB22 tok/s1093K ctx
dense
logos16v2 stablelm2 1.6b i1
C43
1.6B4.7 GB22 tok/s1093K ctx
dense
Yi Coder 1.5B
C43
1.5B4.6 GB21 tok/s1172K ctx
dense
👁 Unsloth
Falcon H1 1.5B Instruct
C42
1.5B4.6 GB21 tok/s1172K ctx
dense
TinyLlama 1.1B Chat v1.0
C42
1.1B4.3 GB15 tok/s1628K ctx
dense
EXAONE 4.0 1.2B
C42
1.2B4.4 GB17 tok/s1485K ctx
dense
Llama 3.2 1B Instruct Q8 0
C42
1B4.4 GB14 tok/s1771K ctx
dense
TinyLlama 1.1B Chat v0.3
C41
1.1B4.3 GB15 tok/s1628K ctx
dense
👁 TinyLlama
TinyLlama 1.1B Chat v0.6
C41
1.1B4.3 GB15 tok/s1628K ctx
dense
👁 Alibaba
Qwen 2.5 0.5B
C41
0.5B4.0 GB7 tok/s131K ctx
dense
Llama 3.2 1B Instruct
C41
1B4.2 GB14 tok/s1799K ctx
dense
gemma 3 1b it
C41
1B4.2 GB14 tok/s1799K ctx
dense
TinyLlama 1.1B Chat v1.0 imatrix
C41
1.1B4.3 GB15 tok/s1628K ctx
dense
Falcon3 1B Instruct abliterated
C41
1B4.2 GB14 tok/s1799K ctx
dense
F

F Tier -- Can't Run (117 models)

Too heavy for this hardware.

SmolVLM 500M Instruct
D39
0.5B4.0 GB7 tok/s2140K ctx
dense
Codestral 22B v0.1
D39
22B19.5 GB18 tok/s4K ctx
dense
Codestral 22B v0.1
D39
22B19.5 GB18 tok/s4K ctx
dense
Codestral 22B v0.1
D38
22B19.5 GB18 tok/s4K ctx
dense
Codestral 21B Pruned i1
D38
21B18.8 GB19 tok/s6K ctx
dense
Codestral 22B v0.1 IMat
D38
22B19.5 GB18 tok/s4K ctx
dense
Codestral 22B v0.1 i1
D38
22B19.5 GB18 tok/s4K ctx
dense
embeddinggemma 300M
D37
0.3B3.8 GB4 tok/s2167K ctx
dense
granite embedding 107m multilingual
D35
0.11B3.7 GB2 tok/s2196K ctx
dense
Falcon H1 Tiny 90M Instruct
D35
0.09B3.6 GB2 tok/s2197K ctx
dense
👁 Meta
Llama 3.1 70B
F0
70B51.1 GB4 tok/s4K ctx
dense
👁 Meta
Llama 3.3 70B
F0
70B51.1 GB4 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 32B
F0
32B26.9 GB12 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 72B
F0
72B52.3 GB4 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 Coder 32B
F0
32B26.9 GB12 tok/s4K ctx
dense
👁 Alibaba
Qwen 3 32B
F0
32B26.9 GB12 tok/s4K ctx
dense
👁 Alibaba
Qwen 3 235B A22B
F0
235B149.7 GB2 tok/s4K ctx
moe
👁 Alibaba
Qwen 3 30B A3B
F0
30.5B23.6 GB21 tok/s4K ctx
moe
👁 Alibaba
Qwen3-Coder 30B A3B Instruct
F0
30.5B23.6 GB21 tok/s4K ctx
moe
👁 Alibaba
Qwen3-Coder-Next
F0
80B53.8 GB7 tok/s4K ctx
moe
👁 Alibaba
Qwen3-Coder 480B A35B Instruct
F0
480B299.2 GB2 tok/s4K ctx
moe
👁 Alibaba
Qwen3-VL 30B A3B Instruct
F0
30B23.3 GB22 tok/s4K ctx
moe
👁 Alibaba
Qwen 3.5 27B
F0
27B23.1 GB15 tok/s4K ctx
dense
👁 Alibaba
Qwen 3.5 35B A3B
F0
35B26.3 GB18 tok/s4K ctx
moe
👁 Alibaba
Qwen 3.6 35B A3B
F0
35B28.9 GB15 tok/s4K ctx
+1moe
👁 Alibaba
Qwen 3.6 27B
F0
27B20.9 GB13 tok/s4K ctx
+1dense
👁 Alibaba
Qwen 3.5 122B A10B
F0
122B80.4 GB6 tok/s4K ctx
moe
👁 Alibaba
Qwen 3.5 397B A17B
F0
397B248.5 GB2 tok/s4K ctx
moe
👁 Google
Gemma 2 27B
F0
27B31.2 GB8 tok/s4K ctx
dense
👁 Google
Gemma 3 27B
F0
27B31.2 GB8 tok/s4K ctx
dense
👁 Mistral
Mixtral 8x7B
F0
47B34.1 GB7 tok/s4K ctx
moe
👁 Mistral
Mixtral 8x22B
F0
141B92.9 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek R1 Distill 32B
F0
32B26.9 GB12 tok/s4K ctx
dense
👁 IBM
Granite 4.1 30B
F0
30B25.7 GB13 tok/s4K ctx
dense
👁 Cohere
Command R 35B
F0
35B27.3 GB11 tok/s4K ctx
dense
👁 TII
Falcon 40B Instruct
F0
40B34.1 GB7 tok/s4K ctx
dense
👁 WizardLM
WizardLM 13B
F0
13B23.6 GB15 tok/s8K ctx
dense
👁 BigCode
StarCoder 15B
F0
15B28.9 GB10 tok/s4K ctx
dense
👁 Meta
CodeLlama 13B Instruct
F0
13B23.6 GB15 tok/s8K ctx
dense
👁 MosaicML
MPT-30B-Instruct
F0
30B48.5 GB8 tok/s4K ctx
dense
👁 Mistral
Devstral 2 123B Instruct
F0
123B83.9 GB3 tok/s4K ctx
dense
👁 Mistral
Mistral Small 4 119B
F0
119B81.5 GB6 tok/s4K ctx
moe
👁 Mistral
Leanstral 119B A6B
F0
119B84.9 GB6 tok/s4K ctx
moe
👁 Mistral
Mistral Large 3
F0
675B420.6 GB2 tok/s4K ctx
+1moe
F0
744B476.4 GB2 tok/s4K ctx
moe
👁 Z.ai
GLM-5.1
F0
754B482.5 GB2 tok/s4K ctx
moe
F0
1000B620.9 GB2 tok/s4K ctx
moe
F0
1000B620.9 GB2 tok/s4K ctx
+1moe
👁 Moonshot AI
Kimi Linear 48B A3B
F0
48B33.7 GB6 tok/s4K ctx
linear
👁 Nous Research
Nous Hermes 1.0
F0
9B21.2 GB26 tok/s11K ctx
dense
👁 Nous Research
Nous Dolphin 13B
F0
13B25.1 GB12 tok/s6K ctx
dense
👁 Baichuan
Baichuan 13B
F0
13B25.1 GB12 tok/s6K ctx
dense
👁 Cerebras
Cerebras-GPT 13B
F0
13B22.6 GB14 tok/s7K ctx
dense
👁 Stability AI
StableLM 2 12B
F0
12B24.3 GB13 tok/s4K ctx
dense
👁 InternLM
InternLM 20B
F0
20B36.2 GB9 tok/s4K ctx
dense
👁 01.AI
Yi 34B Chat
F0
34B27.9 GB11 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek V3 671B
F0
671B472.4 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V3.1 671B
F0
671B472.4 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V4 Pro
F0
1600B867.4 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V4 Flash
F0
284B162.8 GB3 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek R1 671B
F0
671B472.4 GB2 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek Coder V2 236B
F0
236B206.0 GB3 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek V2.5 236B
F0
236B206.0 GB3 tok/s4K ctx
moe
👁 DeepSeek
DeepSeek LLM 67B
F0
67B50.2 GB5 tok/s4K ctx
dense
👁 Meta
Llama 4 Scout 17B 16E
F0
109B72.9 GB4 tok/s4K ctx
moe
👁 Meta
Llama 4 Maverick 17B 128E
F0
400B250.4 GB2 tok/s4K ctx
moe
👁 Cohere
Command R+ 104B
F0
104B70.3 GB3 tok/s4K ctx
dense
👁 Cohere
Command A 111B
F0
111B75.1 GB3 tok/s4K ctx
dense
👁 01.AI
Yi 1.5 34B
F0
34B27.9 GB11 tok/s4K ctx
dense
👁 Mistral AI
Pixtral Large 124B
F0
124B84.5 GB2 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 VL 72B
F0
72B52.3 GB4 tok/s4K ctx
dense
👁 Alibaba
Qwen 2.5 Math 72B
F0
72B52.3 GB4 tok/s4K ctx
dense
👁 NVIDIA
Nemotron 70B
F0
70B51.1 GB5 tok/s4K ctx
dense
👁 LLaVA
LLaVA 1.6 13B
F0
13B23.6 GB15 tok/s4K ctx
dense
👁 LMSYS
Vicuna 13B
F0
13B23.6 GB15 tok/s4K ctx
dense
👁 Allen AI
OLMo 2 32B
F0
32B26.9 GB12 tok/s4K ctx
dense
👁 NVIDIA
Nemotron 3 Nano 30B
F0
30B24.2 GB15 tok/s4K ctx
dense
👁 LG AI
EXAONE 4.0 32B
F0
32B26.9 GB12 tok/s4K ctx
dense
👁 Meta
Llama 3.1 405B
F0
405B258.2 GB2 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek R1 Distill 70B
F0
70B51.1 GB4 tok/s4K ctx
dense
👁 IBM
Granite Code 34B
F0
34B27.9 GB11 tok/s4K ctx
dense
👁 OpenAI
GPT-OSS 120B
F0
117B79.7 GB3 tok/s4K ctx
dense
👁 Cohere
Aya Expanse 32B
F0
32B25.5 GB13 tok/s4K ctx
dense
👁 Google
Gemma 4 26B A4B
F0
25.2B22.5 GB24 tok/s4K ctx
moe
👁 Google
Gemma 4 31B
F0
30.7B36.9 GB7 tok/s4K ctx
dense
👁 DeepSeek
DeepSeek V3.2
F0
671B413.3 GB2 tok/s4K ctx
moe
👁 NVIDIA
Nemotron Cascade 2 30B A3B
F0
30B24.7 GB20 tok/s4K ctx
moe
MiniMax M2.7
F0
230B147.6 GB3 tok/s4K ctx
moe
Solar Open 100B
F0
100B76.2 GB3 tok/s4K ctx
dense
Baichuan M2 32B Q4 K M
F0
32B26.8 GB11 tok/s4K ctx
dense
baichuan inc Baichuan M2 32B
F0
32B26.8 GB11 tok/s4K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
F0
24B20.9 GB16 tok/s4K ctx
dense
cognitivecomputations Dolphin3.0 R1 Mistral 24B
F0
24B20.9 GB16 tok/s4K ctx
dense
Qwen3 48B A4B Savant Commander Distill 12X Closed Open Heretic Uncensored
F0
48B38.4 GB6 tok/s4K ctx
dense
EXAONE 4.0 32B
F0
32B26.8 GB11 tok/s4K ctx
dense
K EXAONE 236B A23B
F0
236B175.1 GB2 tok/s4K ctx
dense
Qwen3.5 35B A3B
F0
35B28.9 GB9 tok/s4K ctx
dense
gemma 3 27b it
F0
27B23.1 GB14 tok/s4K ctx
dense
Llama 3.3 70B Instruct
F0
70B54.4 GB4 tok/s4K ctx
dense
Mistral Small 24B Instruct 2501
F0
24B20.9 GB16 tok/s4K ctx
dense
mistral small 3.1 24b instruct 2503 hf
F0
24B20.9 GB16 tok/s4K ctx
dense
aya expanse 32b heretic MPOA i1
F0
32B26.8 GB11 tok/s4K ctx
dense
Baichuan M3 235B
F0
235B174.4 GB2 tok/s4K ctx
dense
Baichuan M3 235B i1
F0
235B174.4 GB2 tok/s4K ctx
dense
BaichuanMed OCR 72B i1
F0
72B55.8 GB4 tok/s4K ctx
dense
Dolphin Mistral GLM 4.7 Flash 24B Venice Edition Thinking Uncensored i1
F0
24B20.9 GB16 tok/s4K ctx
dense
Solar Open 100B i1
F0
100B76.2 GB3 tok/s4K ctx
dense
Solar Open 69B REAP i1
F0
69B53.7 GB4 tok/s4K ctx
dense
Hermes 4.3 36B
F0
36B29.7 GB9 tok/s4K ctx
dense
stabilityai japanese stablelm instruct beta 70b
F0
70B54.4 GB4 tok/s4K ctx
dense
👁 Unsloth
gemma 3 27b it
F0
27B23.1 GB14 tok/s4K ctx
dense
👁 Unsloth
Mistral Small 3.2 24B Instruct 2506
F0
24B20.9 GB16 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 122B A10B
F0
122B77.6 GB3 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 27B
F0
27B23.1 GB14 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 35B A3B
F0
35B28.9 GB9 tok/s4K ctx
dense
👁 Unsloth
Qwen3.5 397B A17B
F0
397B292.2 GB2 tok/s4K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
F0
24B20.9 GB16 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 24GB

24 GB unified memory

View tier list

Mac mini M2 24GB

24 GB unified memory

View tier list

MacBook Air M3 24GB

24 GB unified memory

View tier list

MacBook Air M4 24GB

24 GB unified memory

View tier list
Full MacBook Pro M4 Pro 24GB pageCompare this hardware