AI Compatibility Tier List
MacBook Air M3 24GB
374 AI models ranked by how well they run on your 24 GB unified memory hardware. Best match: Qwen 3.5 9B (score: 91).
374
Models tested
257
Can run
24 GB
unified memory
13
Best tok/s
Tier Distribution
S 5
A 36
B 47
C 169
F 117
S: 5A: 36B: 47C: 169F: 117
S
S Tier -- Perfect Fit (5 models)
These models run flawlessly with headroom to spare.
A
A Tier -- Great Fit (36 models)
Excellent performance with comfortable memory margins.
A85
3.8B7.3 GB32 tok/s125K ctx
dense
A84
8B10.3 GB15 tok/s73K ctx
dense
A81
14B14.5 GB9 tok/s34K ctx
multimodal
A81
8B10.3 GB15 tok/s8K ctx
dense
A80
8B10.6 GB15 tok/s65K ctx
multimodal
A80
4B8.1 GB30 tok/s33K ctx
dense
A79
8B10.3 GB15 tok/s2K ctx
dense
A79
9B11.4 GB13 tok/s54K ctx
dense
A79
0.57B6.6 GB8 tok/s8K ctx
dense
A79
19B17.5 GB6 tok/s8K ctx
dense
A78
7B9.7 GB17 tok/s8K ctx
dense
A78
14B15.1 GB9 tok/s16K ctx
dense
A77
0.57B5.8 GB8 tok/s8K ctx
dense
A77
14B15.0 GB9 tok/s29K ctx
dense
A77
16B16.5 GB17 tok/s20K ctx
moe
A77
7B8.6 GB17 tok/s33K ctx
dense
A77
13B13.9 GB9 tok/s33K ctx
dense
A76
7B9.7 GB17 tok/s8K ctx
dense
A76
21B18.7 GB11 tok/s6K ctx
moe
A76
12B15.7 GB7 tok/s21K ctx
dense
A75
9B9.6 GB14 tok/s131K ctx
dense
A75
8B9.7 GB11 tok/s111K ctx
dense
A74
3B7.5 GB39 tok/s87K ctx
dense
A74
8B10.3 GB15 tok/s8K ctx
dense
A74
0.34B5.6 GB5 tok/s1K ctx
dense
A74
8B10.8 GB15 tok/s58K ctx
dense
A74
7B9.7 GB17 tok/s8K ctx
dense
A73
12B13.3 GB10 tok/s42K ctx
dense
A73
7B8.6 GB17 tok/s131K ctx
dense
A73
0.34B5.6 GB5 tok/s1K ctx
dense
A72
7B8.3 GB18 tok/s262K ctx
state-space
A72
7B15.1 GB16 tok/s8K ctx
dense
A72
7B15.6 GB16 tok/s16K ctx
dense
A71
0.34B5.6 GB5 tok/s1K ctx
dense
A71
14B15.0 GB9 tok/s29K ctx
dense
A70
7B10.7 GB17 tok/s8K ctx
dense
B
B Tier -- Good Fit (47 models)
Solid performance, may need quantization or reduced context.
B70
8B10.3 GB15 tok/s73K ctx
dense
B70
3.8B11.7 GB29 tok/s31K ctx
dense
B70
24B20.6 GB4 tok/s4K ctx
dense
B69
24B20.6 GB4 tok/s4K ctx
dense
B69
3B6.1 GB40 tok/s261K ctx
multimodal
B69
7B9.7 GB17 tok/s4K ctx
dense
B69
7B9.7 GB17 tok/s4K ctx
dense
B69
9B9.6 GB14 tok/s128K ctx
dense
B69
7B15.6 GB16 tok/s8K ctx
dense
B69
7B15.6 GB16 tok/s8K ctx
dense
B69
22B19.4 GB4 tok/s4K ctx
dense
B68
4B8.0 GB22 tok/s87K ctx
dense
B68
4B11.8 GB28 tok/s31K ctx
dense
B68
5.1B7.1 GB18 tok/s128K ctx
dense
B68
2B6.4 GB28 tok/s118K ctx
dense
B68
0.14B4.3 GB2 tok/s8K ctx
dense
B68
4B7.4 GB30 tok/s124K ctx
dense
B68
24B20.6 GB4 tok/s4K ctx
dense
B67
3B7.5 GB39 tok/s87K ctx
dense
B67
7B8.6 GB17 tok/s131K ctx
dense
B66
7B15.6 GB16 tok/s4K ctx
dense
B65
8B10.3 GB15 tok/s33K ctx
dense
B65
1.7B6.2 GB24 tok/s33K ctx
dense
B65
3B7.8 GB37 tok/s8K ctx
dense
B65
7B9.7 GB17 tok/s4K ctx
dense
B65
20B18.9 GB5 tok/s8K ctx
dense
B64
11B12.2 GB11 tok/s16K ctx
vision
B64
7B15.6 GB16 tok/s8K ctx
dense
B64
7B15.6 GB16 tok/s8K ctx
dense
B64
3B6.5 GB40 tok/s131K ctx
dense
B64
9B14.1 GB10 tok/s8K ctx
dense
B64
7B8.6 GB17 tok/s33K ctx
dense
B62
7B7.9 GB18 tok/s8K ctx
dense
B62
12B13.3 GB10 tok/s42K ctx
dense
B62
0.02B3.8 GB2 tok/s0K ctx
dense
B62
24B20.6 GB4 tok/s4K ctx
vision
B62
1.5B4.8 GB21 tok/s33K ctx
dense
B61
24B20.6 GB4 tok/s4K ctx
dense
B61
9B10.4 GB14 tok/s91K ctx
dense
B61
7B9.7 GB17 tok/s8K ctx
dense
B61
14B15.0 GB9 tok/s29K ctx
dense
B60
3B7.0 GB40 tok/s112K ctx
dense
B60
24B20.6 GB4 tok/s4K ctx
dense
B59
24B20.6 GB4 tok/s4K ctx
dense
B59
8B10.6 GB15 tok/s65K ctx
dense
B57
14B15.1 GB9 tok/s28K ctx
dense
B56
3B7.3 GB40 tok/s98K ctx
dense
C
C Tier -- Tight Fit (169 models)
Runs with offloading or significant compromises.
C54
8B10.3 GB17 tok/s73K ctx
state-space
C54
1.5B4.8 GB21 tok/s33K ctx
dense
C53
1.1B4.5 GB15 tok/s4K ctx
dense
C53
9B10.4 GB14 tok/s4K ctx
dense
C51
7B8.6 GB17 tok/s4K ctx
dense
C51
1.5B4.8 GB21 tok/s131K ctx
dense
C51
2B6.3 GB28 tok/s8K ctx
dense
C50
7B9.7 GB17 tok/s8K ctx
dense
C50
8B10.3 GB15 tok/s8K ctx
dense
gemma 3 12b it
C5012B12.2 GB9 tok/s74K ctx
dense
C49
8B10.3 GB15 tok/s33K ctx
dense
C49
1B4.5 GB14 tok/s33K ctx
dense
C49
14B13.7 GB8 tok/s51K ctx
dense
C49
7B9.7 GB17 tok/s8K ctx
dense
C49
7B9.7 GB17 tok/s33K ctx
dense
NousResearch Hermes 4 14B
C4914B13.7 GB8 tok/s51K ctx
dense
SOLAR 10.7B Instruct v1.0 uncensored
C4910.7B11.3 GB10 tok/s93K ctx
dense
Nous Hermes 2 SOLAR 10.7B
C4910.7B11.3 GB10 tok/s93K ctx
dense
C49
4B7.9 GB30 tok/s4K ctx
dense
GGUF SOLARized GraniStral 14B 2102 YeAM HCT 32QKV
C4914B13.7 GB8 tok/s51K ctx
dense
C49
7B15.6 GB16 tok/s4K ctx
dense
C49
7B9.7 GB17 tok/s8K ctx
dense
internlm JanusCoder 14B
C4914B13.7 GB8 tok/s51K ctx
dense
GGUF SOLARized GraniStral 14B 1902 YeAM HCT
C4914B13.7 GB8 tok/s51K ctx
dense
solar finalised finetuned Model 10.7B i1
C4910.7B11.3 GB10 tok/s93K ctx
dense
SOLAR 10.7B v1.0
C4910.7B11.3 GB10 tok/s93K ctx
dense
C48
9B10.0 GB12 tok/s126K ctx
dense
Qwen3.5 9B Uncensored HauhauCS Aggressive
C489B10.0 GB12 tok/s126K ctx
dense
Qwen3.5 9B
C489B10.0 GB12 tok/s126K ctx
dense
Helply 10.2b chat i1
C4810.2B10.9 GB11 tok/s101K ctx
dense
HelpingAI2.5 10B i1
C4810B10.8 GB11 tok/s105K ctx
dense
Yi Coder 9B Chat
C489B10.0 GB12 tok/s126K ctx
dense
glm 4 9b chat 1m
C489B10.0 GB12 tok/s126K ctx
dense
Yi 1.5 9B Chat
C489B10.0 GB12 tok/s126K ctx
dense
C48
7B9.7 GB17 tok/s8K ctx
dense
Meta Llama 3.1 8B Instruct
C488B9.3 GB14 tok/s152K ctx
dense
llava llama 3 8b v1 1
C488B9.3 GB14 tok/s152K ctx
dense
C48
0.5B4.0 GB7 tok/s131K ctx
dense
C48
8B9.3 GB14 tok/s152K ctx
dense
Meta Llama 3 8B Instruct
C488B9.3 GB14 tok/s152K ctx
dense
C48
8B9.3 GB14 tok/s152K ctx
dense
Dolphin3.0 Llama3.1 8B
C488B9.3 GB14 tok/s152K ctx
dense
Llama 3 8B Instruct 32k v0.1
C488B9.3 GB14 tok/s152K ctx
dense
Yi 9B Coder i1
C489B10.0 GB12 tok/s126K ctx
dense
Meta Llama 3.1 8B Instruct
C488B9.3 GB14 tok/s152K ctx
dense
vntl llama3 8b v2
C488B9.3 GB14 tok/s152K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C478B9.3 GB14 tok/s152K ctx
dense
HelpingAI2 9B
C479B10.0 GB12 tok/s126K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C478B9.3 GB14 tok/s152K ctx
dense
Qwen3 8B DeepSeek v3.2 Speciale Distill
C478B9.3 GB14 tok/s152K ctx
dense
HelpingAI2 9B i1
C479B10.0 GB12 tok/s126K ctx
dense
HelpingAI 9B 200k i1
C479B10.0 GB12 tok/s126K ctx
dense
Hermes 3 Llama 3.1 8B
C478B9.3 GB14 tok/s152K ctx
dense
HelpingAI 9B i1
C479B10.0 GB12 tok/s126K ctx
dense
Hermes 2 Pro Llama 3 8B
C478B9.3 GB14 tok/s152K ctx
dense
Dolphin3.0 Llama3.1 8B
C478B9.3 GB14 tok/s152K ctx
dense
granite 8b code instruct 4k
C478B9.3 GB14 tok/s152K ctx
dense
dolphin 2.9.4 llama3.1 8b
C478B9.3 GB14 tok/s152K ctx
dense
Llama 2 7B Chat
C477B8.6 GB16 tok/s186K ctx
dense
falcon mamba 7b instruct Q4 K M
C477B8.6 GB18 tok/s186K ctx
dense
openchat 3.6 8b 20240522 IMat
C478B9.3 GB14 tok/s152K ctx
dense
Mistral 7B Instruct v0.2
C477B8.6 GB16 tok/s186K ctx
dense
aya expanse 8b
C478B9.3 GB14 tok/s152K ctx
dense
C47
15B15.5 GB7 tok/s16K ctx
dense
Mistral 7B Instruct v0.3
C477B8.6 GB16 tok/s186K ctx
dense
EXAONE 3.5 7.8B Instruct
C477.8B9.2 GB14 tok/s158K ctx
dense
dolphin v2 8b abliterated i1
C478B9.3 GB14 tok/s152K ctx
dense
EXAONE 3.5 7.8B Instruct
C477.8B9.2 GB14 tok/s158K ctx
dense
aya expanse 8b orthogonal heretic i1
C478B9.3 GB14 tok/s152K ctx
dense
EXAONE 3.5 7.8B Instruct i1
C477.8B9.2 GB14 tok/s158K ctx
dense
exaone 3.0 7.8b it
C477.8B9.2 GB14 tok/s158K ctx
dense
Mamba Codestral 7B v0.1
C477B8.6 GB18 tok/s186K ctx
dense
aya expanse 8b orthogonal heretic
C478B9.3 GB14 tok/s152K ctx
dense
Mistral 7B Instruct v0.3
C477B8.6 GB16 tok/s186K ctx
dense
Hermes 2 Pro Mistral 7B
C477B8.6 GB16 tok/s186K ctx
dense
internlm2 5 1 8b chat i1
C478B9.3 GB14 tok/s152K ctx
dense
zephyr 7B beta
C477B8.6 GB16 tok/s186K ctx
dense
internlm3 8b instruct abliterated i1
C478B9.3 GB14 tok/s152K ctx
dense
C47
6B8.1 GB20 tok/s4K ctx
dense
Nous Hermes 2 Mistral 7B DPO
C477B8.6 GB16 tok/s186K ctx
dense
Falcon H1R 7B
C477B8.6 GB16 tok/s186K ctx
dense
zephyr 7B alpha
C477B8.6 GB16 tok/s186K ctx
dense
speechless zephyr code functionary 7b
C477B8.6 GB16 tok/s186K ctx
dense
StarCoder2 7B
C477B8.6 GB16 tok/s186K ctx
dense
japanese stablelm instruct gamma 7B
C477B8.6 GB16 tok/s186K ctx
dense
Falcon H1 7B Instruct
C477B8.6 GB16 tok/s186K ctx
dense
C47
7B8.6 GB16 tok/s186K ctx
dense
C47
7B15.1 GB16 tok/s4K ctx
dense
Yi 1.5 6B Chat
C476B7.9 GB19 tok/s230K ctx
dense
stabilityai japanese stablelm base gamma 7b
C477B8.6 GB16 tok/s186K ctx
dense
stablelm 2 zephyr 1 6b
C476B7.9 GB19 tok/s230K ctx
dense
baichuan2 7b chat
C477B8.6 GB16 tok/s186K ctx
dense
DiscoPOP zephyr 7b gemma
C467B8.6 GB16 tok/s186K ctx
dense
starcoder2 7b
C467B8.6 GB16 tok/s186K ctx
dense
zephyr 7b beta Mistral 7B Instruct v0.2
C467B8.6 GB16 tok/s186K ctx
dense
OpenChat 3.5 7B Qwen v2.0 i1
C467B8.6 GB16 tok/s186K ctx
dense
OpenChat 3.5 7B Starling v2.0 i1
C467B8.6 GB16 tok/s186K ctx
dense
internlm2 math plus 7b IMat
C467B8.6 GB16 tok/s186K ctx
dense
CodeNinja 1.0 OpenChat 7B i1
C467B8.6 GB16 tok/s186K ctx
dense
internlm2 5 7b chat i1
C467B8.6 GB16 tok/s186K ctx
dense
OpenSafetyLab MD Judge v0 2 internlm2 7b
C467B8.6 GB16 tok/s186K ctx
dense
MD Judge v0 2 internlm2 7b i1
C467B8.6 GB16 tok/s186K ctx
dense
zephyr 7b gemma sft african ultrachat 100k
C467B8.6 GB16 tok/s186K ctx
dense
jointpreferences mistral 7b sft helpful
C467B8.6 GB16 tok/s186K ctx
dense
zephyr 7b dpo full i1
C467B8.6 GB16 tok/s186K ctx
dense
blossom v3 baichuan2 7b i1
C467B8.6 GB16 tok/s186K ctx
dense
C46
0.6B4.7 GB8 tok/s33K ctx
dense
Yi 1.5 6B Chat
C466B7.9 GB19 tok/s230K ctx
dense
blossom v1 baichuan 7b i1
C467B8.6 GB16 tok/s186K ctx
dense
stablelm 2 1 6b chat imatrix
C466B7.9 GB19 tok/s230K ctx
dense
C46
3B5.7 GB37 tok/s544K ctx
dense
C46
4B6.4 GB28 tok/s387K ctx
dense
Llama 3.2 3B Instruct
C463B6.0 GB32 tok/s529K ctx
dense
gemma 3 4b it
C464B6.4 GB28 tok/s387K ctx
dense
C46
1B4.6 GB14 tok/s128K ctx
dense
Llama 3.2 3B Instruct
C463B5.7 GB37 tok/s544K ctx
dense
Ministral 3 3B Instruct 2512
C463B5.7 GB37 tok/s544K ctx
dense
gemma 3 4b it
C464B6.4 GB28 tok/s387K ctx
dense
HelpingAI2 6B
C466B7.9 GB19 tok/s230K ctx
dense
C46
7B8.3 GB17 tok/s16K ctx
dense
HelpingAI2 6B i1
C466B7.9 GB19 tok/s230K ctx
dense
HELVETE 3B
C463B5.7 GB37 tok/s544K ctx
dense
Hermes 3 Llama 3.2 3B
C463B5.7 GB37 tok/s544K ctx
dense
stablelm zephyr 3b
C463B5.7 GB37 tok/s544K ctx
dense
Phi 4 reasoning vision 15B
C4615B14.4 GB7 tok/s42K ctx
dense
StarCoder2 3B
C463B5.7 GB37 tok/s544K ctx
dense
StarCoder2 15B
C4615B14.4 GB7 tok/s42K ctx
dense
AI21 Jamba Reasoning 3B
C463B5.7 GB37 tok/s544K ctx
dense
stablelm 3b 4e1t
C463B5.7 GB37 tok/s544K ctx
dense
ai21labs AI21 Jamba Reasoning 3B
C463B5.7 GB37 tok/s544K ctx
dense
HelpingAI2.5 5B i1
C465B7.1 GB22 tok/s293K ctx
dense
starcoder2 15b instruct v0.1
C4615B14.4 GB7 tok/s42K ctx
dense
starcoder2 15b i1
C4615B14.4 GB7 tok/s42K ctx
dense
ai21labs AI21 Jamba2 3B
C463B5.7 GB37 tok/s544K ctx
dense
starcoder2 15b instruct v0.1
C4615B14.4 GB7 tok/s42K ctx
dense
HelpingAI 3B hindi i1
C463B5.7 GB37 tok/s544K ctx
dense
AI21 Jamba2 3B
C463B5.7 GB37 tok/s544K ctx
dense
HelpingAI 3B hindi
C453B5.7 GB37 tok/s544K ctx
dense
HelpingAI 15B i1
C4515B14.4 GB7 tok/s42K ctx
dense
AI21 Jamba2 3B i1
C453B5.7 GB37 tok/s544K ctx
dense
gemma 2 2b it
C452B5.4 GB28 tok/s829K ctx
dense
C45
3B5.8 GB40 tok/s16K ctx
dense
EXAONE 3.5 2.4B Instruct
C452.4B5.2 GB34 tok/s701K ctx
dense
Codestral RAG 19B Pruned i1
C4519B17.3 GB6 tok/s16K ctx
dense
C45
22B19.4 GB5 tok/s4K ctx
dense
C44
2B4.9 GB28 tok/s858K ctx
dense
internlm2 5 20b chat
C4420B18.0 GB5 tok/s11K ctx
dense
internlm2 limarp chat 20b
C4420B18.0 GB5 tok/s11K ctx
dense
internlm2 math plus 20b i1
C4420B18.0 GB5 tok/s11K ctx
dense
Gemmasutra Mini 2B v1
C442B4.9 GB28 tok/s858K ctx
dense
gemma 2 2b it
C442B4.9 GB28 tok/s858K ctx
dense
C44
0.6B4.7 GB8 tok/s131K ctx
dense
C43
1.5B4.6 GB21 tok/s1172K ctx
dense
C43
1.5B4.6 GB21 tok/s1172K ctx
dense
Yi Coder 1.5B Chat
C431.5B4.6 GB21 tok/s1172K ctx
dense
stablelm 2 zephyr 1.6b
C431.6B4.7 GB22 tok/s1093K ctx
dense
logos16v2 stablelm2 1.6b i1
C431.6B4.7 GB22 tok/s1093K ctx
dense
Yi Coder 1.5B
C431.5B4.6 GB21 tok/s1172K ctx
dense
C42
1.5B4.6 GB21 tok/s1172K ctx
dense
TinyLlama 1.1B Chat v1.0
C421.1B4.3 GB15 tok/s1628K ctx
dense
EXAONE 4.0 1.2B
C421.2B4.4 GB17 tok/s1485K ctx
dense
Llama 3.2 1B Instruct Q8 0
C421B4.4 GB14 tok/s1771K ctx
dense
TinyLlama 1.1B Chat v0.3
C411.1B4.3 GB15 tok/s1628K ctx
dense
C41
1.1B4.3 GB15 tok/s1628K ctx
dense
C41
0.5B4.0 GB7 tok/s131K ctx
dense
Llama 3.2 1B Instruct
C411B4.2 GB14 tok/s1799K ctx
dense
gemma 3 1b it
C411B4.2 GB14 tok/s1799K ctx
dense
TinyLlama 1.1B Chat v1.0 imatrix
C411.1B4.3 GB15 tok/s1628K ctx
dense
Falcon3 1B Instruct abliterated
C411B4.2 GB14 tok/s1799K ctx
dense
F
F Tier -- Can't Run (117 models)
Too heavy for this hardware.
SmolVLM 500M Instruct
D390.5B4.0 GB7 tok/s2140K ctx
dense
embeddinggemma 300M
D370.3B3.8 GB4 tok/s2167K ctx
dense
granite embedding 107m multilingual
D350.11B3.7 GB2 tok/s2196K ctx
dense
Falcon H1 Tiny 90M Instruct
D350.09B3.6 GB2 tok/s2197K ctx
dense
Codestral 22B v0.1
D3422B19.5 GB4 tok/s4K ctx
dense
Codestral 22B v0.1
D3422B19.5 GB4 tok/s4K ctx
dense
Codestral 21B Pruned i1
D3421B18.8 GB5 tok/s6K ctx
dense
Codestral 22B v0.1
D3422B19.5 GB4 tok/s4K ctx
dense
Codestral 22B v0.1 IMat
D3422B19.5 GB4 tok/s4K ctx
dense
Codestral 22B v0.1 i1
D3422B19.5 GB4 tok/s4K ctx
dense
F0
70B51.1 GB2 tok/s4K ctx
dense
F0
70B51.1 GB2 tok/s4K ctx
dense
F0
32B26.9 GB2 tok/s4K ctx
dense
F0
72B52.3 GB2 tok/s4K ctx
dense
F0
32B26.9 GB2 tok/s4K ctx
dense
F0
32B26.9 GB2 tok/s4K ctx
dense
F0
235B149.7 GB2 tok/s4K ctx
moe
F0
30.5B23.6 GB7 tok/s4K ctx
moe
F0
30.5B23.6 GB7 tok/s4K ctx
moe
F0
80B53.8 GB2 tok/s4K ctx
moe
F0
480B299.2 GB2 tok/s4K ctx
moe
F0
30B23.3 GB7 tok/s4K ctx
moe
F0
27B23.1 GB3 tok/s4K ctx
dense
F0
35B26.3 GB6 tok/s4K ctx
moe
F0
35B28.9 GB5 tok/s4K ctx
+1moe
F0
27B20.9 GB3 tok/s4K ctx
+1dense
F0
122B80.4 GB2 tok/s4K ctx
moe
F0
397B248.5 GB2 tok/s4K ctx
moe
F0
27B31.2 GB2 tok/s4K ctx
dense
F0
27B31.2 GB2 tok/s4K ctx
dense
F0
47B34.1 GB2 tok/s4K ctx
moe
F0
141B92.9 GB2 tok/s4K ctx
moe
F0
32B26.9 GB2 tok/s4K ctx
dense
F0
30B25.7 GB2 tok/s4K ctx
dense
F0
35B27.3 GB2 tok/s4K ctx
dense
F0
40B34.1 GB2 tok/s4K ctx
dense
F0
13B23.6 GB6 tok/s8K ctx
dense
F0
15B28.9 GB3 tok/s4K ctx
dense
F0
13B23.6 GB6 tok/s8K ctx
dense
F0
30B48.5 GB2 tok/s4K ctx
dense
F0
123B83.9 GB2 tok/s4K ctx
dense
F0
119B81.5 GB2 tok/s4K ctx
moe
F0
119B84.9 GB2 tok/s4K ctx
moe
F0
675B420.6 GB2 tok/s4K ctx
+1moe
F0
744B476.4 GB2 tok/s4K ctx
moe
F0
754B482.5 GB2 tok/s4K ctx
moe
F0
1000B620.9 GB2 tok/s4K ctx
moe
F0
1000B620.9 GB2 tok/s4K ctx
+1moe
F0
48B33.7 GB2 tok/s4K ctx
linear
F0
9B21.2 GB9 tok/s11K ctx
dense
F0
13B25.1 GB5 tok/s6K ctx
dense
F0
13B25.1 GB5 tok/s6K ctx
dense
F0
13B22.6 GB5 tok/s7K ctx
dense
F0
12B24.3 GB5 tok/s4K ctx
dense
F0
20B36.2 GB2 tok/s4K ctx
dense
F0
34B27.9 GB2 tok/s4K ctx
dense
F0
671B472.4 GB2 tok/s4K ctx
moe
F0
671B472.4 GB2 tok/s4K ctx
moe
F0
1600B867.4 GB2 tok/s4K ctx
moe
F0
284B162.8 GB2 tok/s4K ctx
moe
F0
671B472.4 GB2 tok/s4K ctx
moe
F0
236B206.0 GB2 tok/s4K ctx
moe
F0
236B206.0 GB2 tok/s4K ctx
moe
F0
67B50.2 GB2 tok/s4K ctx
dense
F0
109B72.9 GB2 tok/s4K ctx
moe
F0
400B250.4 GB2 tok/s4K ctx
moe
F0
104B70.3 GB2 tok/s4K ctx
dense
F0
111B75.1 GB2 tok/s4K ctx
dense
F0
34B27.9 GB2 tok/s4K ctx
dense
F0
124B84.5 GB2 tok/s4K ctx
dense
F0
72B52.3 GB2 tok/s4K ctx
dense
F0
72B52.3 GB2 tok/s4K ctx
dense
F0
70B51.1 GB2 tok/s4K ctx
dense
F0
13B23.6 GB6 tok/s4K ctx
dense
F0
13B23.6 GB6 tok/s4K ctx
dense
F0
32B26.9 GB2 tok/s4K ctx
dense
F0
30B24.2 GB3 tok/s4K ctx
dense
F0
32B26.9 GB2 tok/s4K ctx
dense
F0
405B258.2 GB2 tok/s4K ctx
dense
F0
70B51.1 GB2 tok/s4K ctx
dense
F0
34B27.9 GB2 tok/s4K ctx
dense
F0
117B79.7 GB2 tok/s4K ctx
dense
F0
32B25.5 GB2 tok/s4K ctx
dense
F0
25.2B22.5 GB8 tok/s4K ctx
moe
F0
30.7B36.9 GB2 tok/s4K ctx
dense
F0
671B413.3 GB2 tok/s4K ctx
moe
F0
30B24.7 GB7 tok/s4K ctx
moe
MiniMax M2.7
F0230B147.6 GB2 tok/s4K ctx
moe
Solar Open 100B
F0100B76.2 GB2 tok/s4K ctx
dense
Baichuan M2 32B Q4 K M
F032B26.8 GB2 tok/s4K ctx
dense
baichuan inc Baichuan M2 32B
F032B26.8 GB2 tok/s4K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
F024B20.9 GB4 tok/s4K ctx
dense
cognitivecomputations Dolphin3.0 R1 Mistral 24B
F024B20.9 GB4 tok/s4K ctx
dense
Qwen3 48B A4B Savant Commander Distill 12X Closed Open Heretic Uncensored
F048B38.4 GB2 tok/s4K ctx
dense
EXAONE 4.0 32B
F032B26.8 GB2 tok/s4K ctx
dense
K EXAONE 236B A23B
F0236B175.1 GB2 tok/s4K ctx
dense
Qwen3.5 35B A3B
F035B28.9 GB2 tok/s4K ctx
dense
gemma 3 27b it
F027B23.1 GB3 tok/s4K ctx
dense
Llama 3.3 70B Instruct
F070B54.4 GB2 tok/s4K ctx
dense
Mistral Small 24B Instruct 2501
F024B20.9 GB4 tok/s4K ctx
dense
mistral small 3.1 24b instruct 2503 hf
F024B20.9 GB4 tok/s4K ctx
dense
aya expanse 32b heretic MPOA i1
F032B26.8 GB2 tok/s4K ctx
dense
Baichuan M3 235B
F0235B174.4 GB2 tok/s4K ctx
dense
Baichuan M3 235B i1
F0235B174.4 GB2 tok/s4K ctx
dense
BaichuanMed OCR 72B i1
F072B55.8 GB2 tok/s4K ctx
dense
Dolphin Mistral GLM 4.7 Flash 24B Venice Edition Thinking Uncensored i1
F024B20.9 GB4 tok/s4K ctx
dense
Solar Open 100B i1
F0100B76.2 GB2 tok/s4K ctx
dense
Solar Open 69B REAP i1
F069B53.7 GB2 tok/s4K ctx
dense
Hermes 4.3 36B
F036B29.7 GB2 tok/s4K ctx
dense
stabilityai japanese stablelm instruct beta 70b
F070B54.4 GB2 tok/s4K ctx
dense
F0
27B23.1 GB3 tok/s4K ctx
dense
F0
24B20.9 GB4 tok/s4K ctx
dense
F0
122B77.6 GB2 tok/s4K ctx
dense
F0
27B23.1 GB3 tok/s4K ctx
dense
F0
35B28.9 GB2 tok/s4K ctx
dense
F0
397B292.2 GB2 tok/s4K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
F024B20.9 GB4 tok/s4K ctx
dense
Compare with...
See How Other Hardware Stacks Up
Check tier lists for similar GPUs and Apple Silicon configs
