AI Compatibility Tier List
RTX 5070 Ti 16GB
374 AI models ranked by how well they run on your 16 GB VRAM hardware. Best match: Qwen 3.5 9B (score: 97).
374
Models tested
247
Can run
16 GB
VRAM
112
Best tok/s
Tier Distribution
S 11
A 49
B 36
C 151
F 127
S: 11A: 49B: 36C: 151F: 127
S
S Tier -- Perfect Fit (11 models)
These models run flawlessly with headroom to spare.
S97
9B10.2 GB112 tok/s58K ctx
dense
S95
8B9.6 GB126 tok/s63K ctx
dense
S94
14B13.5 GB73 tok/s33K ctx
dense
S93
14.7B14.5 GB66 tok/s24K ctx
dense
S91
4B7.1 GB76 tok/s81K ctx
dense
S89
8B9.3 GB126 tok/s71K ctx
dense
S88
14B13.5 GB70 tok/s33K ctx
multimodal
S87
8B9.3 GB126 tok/s8K ctx
dense
S87
3.8B6.3 GB72 tok/s122K ctx
dense
S85
8B9.6 GB126 tok/s63K ctx
multimodal
S85
9B10.4 GB112 tok/s52K ctx
dense
A
A Tier -- Great Fit (49 models)
Excellent performance with comfortable memory margins.
A85
8B9.3 GB126 tok/s2K ctx
dense
A85
14B14.1 GB70 tok/s16K ctx
dense
A84
19B16.5 GB36 tok/s8K ctx
dense
A84
14B14.0 GB70 tok/s27K ctx
dense
A83
13B12.9 GB76 tok/s33K ctx
dense
A83
7B8.7 GB133 tok/s8K ctx
dense
A82
16B15.6 GB126 tok/s18K ctx
moe
A82
4B7.1 GB76 tok/s33K ctx
dense
A82
12B14.7 GB55 tok/s20K ctx
dense
A82
7B7.6 GB133 tok/s33K ctx
dense
A81
7B8.7 GB133 tok/s8K ctx
dense
A81
9B8.6 GB114 tok/s131K ctx
dense
A81
21B17.8 GB61 tok/s5K ctx
moe
A81
8B8.7 GB87 tok/s108K ctx
dense
A80
12B12.3 GB82 tok/s41K ctx
dense
A79
8B9.3 GB126 tok/s8K ctx
dense
A79
0.57B5.6 GB11 tok/s8K ctx
dense
A79
8B9.8 GB126 tok/s56K ctx
dense
A79
7B8.7 GB133 tok/s8K ctx
dense
A78
7B7.6 GB133 tok/s131K ctx
dense
A78
14B14.0 GB70 tok/s27K ctx
dense
A77
0.57B4.8 GB11 tok/s8K ctx
dense
A77
7B14.1 GB133 tok/s8K ctx
dense
A77
7B14.6 GB133 tok/s16K ctx
dense
A77
7B7.3 GB133 tok/s262K ctx
state-space
A75
7B9.7 GB133 tok/s8K ctx
dense
A75
8B9.3 GB126 tok/s71K ctx
dense
A75
9B8.6 GB114 tok/s128K ctx
dense
A75
3B6.5 GB57 tok/s85K ctx
dense
A75
7B14.6 GB133 tok/s8K ctx
dense
A75
7B14.6 GB133 tok/s8K ctx
dense
A74
7B8.7 GB133 tok/s4K ctx
dense
A74
7B8.7 GB133 tok/s4K ctx
dense
A74
0.34B4.7 GB6 tok/s1K ctx
dense
A73
22B18.4 GB19 tok/s4K ctx
dense
A73
0.34B4.7 GB6 tok/s1K ctx
dense
A73
5.1B6.1 GB97 tok/s128K ctx
dense
A72
3.8B10.7 GB72 tok/s31K ctx
dense
A72
7B7.6 GB133 tok/s131K ctx
dense
A72
4B7.0 GB76 tok/s85K ctx
dense
A72
7B14.6 GB133 tok/s4K ctx
dense
A71
4B10.8 GB76 tok/s30K ctx
dense
A71
0.34B4.7 GB6 tok/s1K ctx
dense
A71
11B11.2 GB89 tok/s16K ctx
vision
A71
8B9.3 GB126 tok/s33K ctx
dense
A71
9B13.1 GB76 tok/s8K ctx
dense
A70
4B6.4 GB76 tok/s121K ctx
dense
A70
7B14.6 GB133 tok/s8K ctx
dense
A70
7B14.6 GB133 tok/s8K ctx
dense
B
B Tier -- Good Fit (36 models)
Solid performance, may need quantization or reduced context.
B70
20B17.9 GB28 tok/s7K ctx
dense
B70
7B8.7 GB133 tok/s4K ctx
dense
B70
3B5.1 GB57 tok/s255K ctx
multimodal
B69
7B7.6 GB133 tok/s33K ctx
dense
B69
12B12.3 GB82 tok/s41K ctx
dense
B69
2B5.4 GB38 tok/s115K ctx
dense
B68
3B6.5 GB57 tok/s85K ctx
dense
B68
0.14B3.3 GB3 tok/s8K ctx
dense
B67
14B14.0 GB70 tok/s27K ctx
dense
B67
7B6.9 GB133 tok/s8K ctx
dense
B67
9B9.5 GB114 tok/s87K ctx
dense
B66
7B8.7 GB133 tok/s8K ctx
dense
B66
3B6.8 GB57 tok/s8K ctx
dense
B65
1.7B5.2 GB32 tok/s33K ctx
dense
B65
3B5.6 GB57 tok/s131K ctx
dense
B64
8B9.6 GB126 tok/s63K ctx
dense
B64
14B14.1 GB70 tok/s26K ctx
dense
B62
1.5B3.8 GB29 tok/s33K ctx
dense
B61
0.02B2.8 GB2 tok/s0K ctx
dense
B61
3B6.0 GB57 tok/s109K ctx
dense
B59
9B9.5 GB114 tok/s4K ctx
dense
B59
8B9.3 GB145 tok/s71K ctx
state-space
B57
3B6.3 GB57 tok/s96K ctx
dense
gemma 3 12b it
B5612B11.2 GB76 tok/s70K ctx
dense
B56
7B7.6 GB133 tok/s4K ctx
dense
B56
14B12.7 GB65 tok/s48K ctx
dense
B56
8B9.3 GB126 tok/s8K ctx
dense
NousResearch Hermes 4 14B
B5614B12.7 GB65 tok/s48K ctx
dense
B56
7B8.7 GB133 tok/s8K ctx
dense
GGUF SOLARized GraniStral 14B 2102 YeAM HCT 32QKV
B5514B12.7 GB65 tok/s48K ctx
dense
SOLAR 10.7B Instruct v1.0 uncensored
B5510.7B10.3 GB85 tok/s89K ctx
dense
internlm JanusCoder 14B
B5514B12.7 GB65 tok/s48K ctx
dense
Nous Hermes 2 SOLAR 10.7B
B5510.7B10.3 GB85 tok/s89K ctx
dense
GGUF SOLARized GraniStral 14B 1902 YeAM HCT
B5514B12.7 GB65 tok/s48K ctx
dense
solar finalised finetuned Model 10.7B i1
B5510.7B10.3 GB85 tok/s89K ctx
dense
SOLAR 10.7B v1.0
B5510.7B10.3 GB85 tok/s89K ctx
dense
C
C Tier -- Tight Fit (151 models)
Runs with offloading or significant compromises.
C55
8B9.3 GB126 tok/s33K ctx
dense
Helply 10.2b chat i1
C5510.2B9.9 GB89 tok/s97K ctx
dense
HelpingAI2.5 10B i1
C5410B9.8 GB94 tok/s101K ctx
dense
C54
7B14.6 GB133 tok/s4K ctx
dense
C54
7B8.7 GB133 tok/s8K ctx
dense
C54
9B9.0 GB105 tok/s122K ctx
dense
Qwen3.5 9B Uncensored HauhauCS Aggressive
C549B9.0 GB105 tok/s122K ctx
dense
Qwen3.5 9B
C549B9.0 GB105 tok/s122K ctx
dense
C54
7B8.7 GB133 tok/s33K ctx
dense
Yi Coder 9B Chat
C549B9.0 GB105 tok/s122K ctx
dense
glm 4 9b chat 1m
C549B9.0 GB105 tok/s122K ctx
dense
C54
1.5B3.8 GB29 tok/s33K ctx
dense
C54
7B8.7 GB133 tok/s8K ctx
dense
Yi 1.5 9B Chat
C549B9.0 GB105 tok/s122K ctx
dense
C54
15B14.5 GB56 tok/s16K ctx
dense
Yi 9B Coder i1
C549B9.0 GB105 tok/s122K ctx
dense
HelpingAI2 9B
C549B9.0 GB105 tok/s122K ctx
dense
HelpingAI2 9B i1
C539B9.0 GB105 tok/s122K ctx
dense
C53
1.1B3.5 GB21 tok/s4K ctx
dense
HelpingAI 9B 200k i1
C539B9.0 GB105 tok/s122K ctx
dense
HelpingAI 9B i1
C539B9.0 GB105 tok/s122K ctx
dense
Meta Llama 3.1 8B Instruct
C538B8.3 GB118 tok/s147K ctx
dense
llava llama 3 8b v1 1
C538B8.3 GB118 tok/s147K ctx
dense
C53
8B8.3 GB118 tok/s147K ctx
dense
Meta Llama 3 8B Instruct
C538B8.3 GB118 tok/s147K ctx
dense
C53
8B8.3 GB118 tok/s147K ctx
dense
Dolphin3.0 Llama3.1 8B
C538B8.3 GB118 tok/s147K ctx
dense
Llama 3 8B Instruct 32k v0.1
C538B8.3 GB118 tok/s147K ctx
dense
Meta Llama 3.1 8B Instruct
C538B8.3 GB118 tok/s147K ctx
dense
vntl llama3 8b v2
C538B8.3 GB118 tok/s147K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C538B8.3 GB118 tok/s147K ctx
dense
DeepSeek R1 0528 Qwen3 8B
C538B8.3 GB118 tok/s147K ctx
dense
Qwen3 8B DeepSeek v3.2 Speciale Distill
C538B8.3 GB118 tok/s147K ctx
dense
Hermes 3 Llama 3.1 8B
C538B8.3 GB118 tok/s147K ctx
dense
Hermes 2 Pro Llama 3 8B
C538B8.3 GB118 tok/s147K ctx
dense
Dolphin3.0 Llama3.1 8B
C538B8.3 GB118 tok/s147K ctx
dense
granite 8b code instruct 4k
C538B8.3 GB118 tok/s147K ctx
dense
dolphin 2.9.4 llama3.1 8b
C538B8.3 GB118 tok/s147K ctx
dense
openchat 3.6 8b 20240522 IMat
C538B8.3 GB118 tok/s147K ctx
dense
C53
7B8.7 GB133 tok/s8K ctx
dense
aya expanse 8b
C538B8.3 GB118 tok/s147K ctx
dense
dolphin v2 8b abliterated i1
C538B8.3 GB118 tok/s147K ctx
dense
aya expanse 8b orthogonal heretic i1
C538B8.3 GB118 tok/s147K ctx
dense
EXAONE 3.5 7.8B Instruct
C537.8B8.2 GB121 tok/s153K ctx
dense
EXAONE 3.5 7.8B Instruct
C537.8B8.2 GB121 tok/s153K ctx
dense
aya expanse 8b orthogonal heretic
C538B8.3 GB118 tok/s147K ctx
dense
EXAONE 3.5 7.8B Instruct i1
C537.8B8.2 GB121 tok/s153K ctx
dense
exaone 3.0 7.8b it
C537.8B8.2 GB121 tok/s153K ctx
dense
internlm2 5 1 8b chat i1
C528B8.3 GB118 tok/s147K ctx
dense
internlm3 8b instruct abliterated i1
C528B8.3 GB118 tok/s147K ctx
dense
Phi 4 reasoning vision 15B
C5215B13.4 GB59 tok/s40K ctx
dense
Llama 2 7B Chat
C527B7.6 GB133 tok/s180K ctx
dense
StarCoder2 15B
C5215B13.4 GB59 tok/s40K ctx
dense
Mistral 7B Instruct v0.2
C527B7.6 GB133 tok/s180K ctx
dense
Mistral 7B Instruct v0.3
C527B7.6 GB133 tok/s180K ctx
dense
C52
7B14.1 GB133 tok/s4K ctx
dense
starcoder2 15b instruct v0.1
C5215B13.4 GB59 tok/s40K ctx
dense
starcoder2 15b i1
C5215B13.4 GB59 tok/s40K ctx
dense
starcoder2 15b instruct v0.1
C5215B13.4 GB59 tok/s40K ctx
dense
Mistral 7B Instruct v0.3
C527B7.6 GB133 tok/s180K ctx
dense
Hermes 2 Pro Mistral 7B
C527B7.6 GB133 tok/s180K ctx
dense
HelpingAI 15B i1
C5215B13.4 GB59 tok/s40K ctx
dense
zephyr 7B beta
C527B7.6 GB133 tok/s180K ctx
dense
Nous Hermes 2 Mistral 7B DPO
C527B7.6 GB133 tok/s180K ctx
dense
Falcon H1R 7B
C527B7.6 GB133 tok/s180K ctx
dense
falcon mamba 7b instruct Q4 K M
C527B7.6 GB133 tok/s180K ctx
dense
zephyr 7B alpha
C527B7.6 GB133 tok/s180K ctx
dense
speechless zephyr code functionary 7b
C527B7.6 GB133 tok/s180K ctx
dense
StarCoder2 7B
C527B7.6 GB133 tok/s180K ctx
dense
japanese stablelm instruct gamma 7B
C527B7.6 GB133 tok/s180K ctx
dense
Falcon H1 7B Instruct
C527B7.6 GB133 tok/s180K ctx
dense
C52
7B7.6 GB133 tok/s180K ctx
dense
Mamba Codestral 7B v0.1
C527B7.6 GB133 tok/s180K ctx
dense
stabilityai japanese stablelm base gamma 7b
C527B7.6 GB133 tok/s180K ctx
dense
baichuan2 7b chat
C527B7.6 GB133 tok/s180K ctx
dense
DiscoPOP zephyr 7b gemma
C527B7.6 GB133 tok/s180K ctx
dense
starcoder2 7b
C527B7.6 GB133 tok/s180K ctx
dense
zephyr 7b beta Mistral 7B Instruct v0.2
C527B7.6 GB133 tok/s180K ctx
dense
OpenChat 3.5 7B Qwen v2.0 i1
C527B7.6 GB133 tok/s180K ctx
dense
OpenChat 3.5 7B Starling v2.0 i1
C527B7.6 GB133 tok/s180K ctx
dense
internlm2 math plus 7b IMat
C527B7.6 GB133 tok/s180K ctx
dense
CodeNinja 1.0 OpenChat 7B i1
C527B7.6 GB133 tok/s180K ctx
dense
internlm2 5 7b chat i1
C527B7.6 GB133 tok/s180K ctx
dense
OpenSafetyLab MD Judge v0 2 internlm2 7b
C527B7.6 GB133 tok/s180K ctx
dense
MD Judge v0 2 internlm2 7b i1
C527B7.6 GB133 tok/s180K ctx
dense
zephyr 7b gemma sft african ultrachat 100k
C527B7.6 GB133 tok/s180K ctx
dense
jointpreferences mistral 7b sft helpful
C527B7.6 GB133 tok/s180K ctx
dense
zephyr 7b dpo full i1
C527B7.6 GB133 tok/s180K ctx
dense
blossom v3 baichuan2 7b i1
C527B7.6 GB133 tok/s180K ctx
dense
blossom v1 baichuan 7b i1
C517B7.6 GB133 tok/s180K ctx
dense
C51
1.5B3.8 GB29 tok/s131K ctx
dense
C51
4B6.9 GB76 tok/s4K ctx
dense
Yi 1.5 6B Chat
C516B6.9 GB114 tok/s224K ctx
dense
C51
6B7.1 GB114 tok/s4K ctx
dense
stablelm 2 zephyr 1 6b
C516B6.9 GB114 tok/s224K ctx
dense
Yi 1.5 6B Chat
C516B6.9 GB114 tok/s224K ctx
dense
stablelm 2 1 6b chat imatrix
C516B6.9 GB114 tok/s224K ctx
dense
C51
7B7.3 GB133 tok/s16K ctx
dense
C51
2B5.3 GB38 tok/s8K ctx
dense
HelpingAI2 6B
C516B6.9 GB114 tok/s224K ctx
dense
HelpingAI2 6B i1
C516B6.9 GB114 tok/s224K ctx
dense
Codestral RAG 19B Pruned i1
C5019B16.3 GB34 tok/s14K ctx
dense
HelpingAI2.5 5B i1
C505B6.1 GB95 tok/s285K ctx
dense
C49
1B3.5 GB19 tok/s33K ctx
dense
C49
22B18.4 GB20 tok/s4K ctx
dense
C49
4B5.4 GB76 tok/s378K ctx
dense
gemma 3 4b it
C494B5.4 GB76 tok/s378K ctx
dense
gemma 3 4b it
C494B5.4 GB76 tok/s378K ctx
dense
C48
0.5B3.0 GB10 tok/s131K ctx
dense
Llama 3.2 3B Instruct
C473B5.0 GB57 tok/s516K ctx
dense
C47
3B4.7 GB57 tok/s531K ctx
dense
Llama 3.2 3B Instruct
C473B4.7 GB57 tok/s531K ctx
dense
Ministral 3 3B Instruct 2512
C473B4.7 GB57 tok/s531K ctx
dense
HELVETE 3B
C473B4.7 GB57 tok/s531K ctx
dense
Hermes 3 Llama 3.2 3B
C463B4.7 GB57 tok/s531K ctx
dense
stablelm zephyr 3b
C463B4.7 GB57 tok/s531K ctx
dense
StarCoder2 3B
C463B4.7 GB57 tok/s531K ctx
dense
C46
0.6B3.7 GB11 tok/s33K ctx
dense
AI21 Jamba Reasoning 3B
C463B4.7 GB57 tok/s531K ctx
dense
stablelm 3b 4e1t
C463B4.7 GB57 tok/s531K ctx
dense
ai21labs AI21 Jamba Reasoning 3B
C463B4.7 GB57 tok/s531K ctx
dense
C46
1B3.6 GB19 tok/s128K ctx
dense
ai21labs AI21 Jamba2 3B
C463B4.7 GB57 tok/s531K ctx
dense
HelpingAI 3B hindi i1
C463B4.7 GB57 tok/s531K ctx
dense
AI21 Jamba2 3B
C463B4.7 GB57 tok/s531K ctx
dense
HelpingAI 3B hindi
C463B4.7 GB57 tok/s531K ctx
dense
AI21 Jamba2 3B i1
C463B4.7 GB57 tok/s531K ctx
dense
C45
3B4.8 GB57 tok/s16K ctx
dense
gemma 2 2b it
C452B4.4 GB38 tok/s810K ctx
dense
EXAONE 3.5 2.4B Instruct
C452.4B4.2 GB46 tok/s685K ctx
dense
C45
2B4.0 GB38 tok/s838K ctx
dense
Gemmasutra Mini 2B v1
C452B4.0 GB38 tok/s838K ctx
dense
gemma 2 2b it
C442B4.0 GB38 tok/s838K ctx
dense
C44
0.6B3.7 GB11 tok/s131K ctx
dense
C43
1.5B3.6 GB29 tok/s1146K ctx
dense
C43
1.5B3.6 GB29 tok/s1146K ctx
dense
Yi Coder 1.5B Chat
C431.5B3.6 GB29 tok/s1146K ctx
dense
stablelm 2 zephyr 1.6b
C431.6B3.7 GB30 tok/s1069K ctx
dense
logos16v2 stablelm2 1.6b i1
C431.6B3.7 GB30 tok/s1069K ctx
dense
Yi Coder 1.5B
C431.5B3.6 GB29 tok/s1146K ctx
dense
C43
1.5B3.6 GB29 tok/s1146K ctx
dense
TinyLlama 1.1B Chat v1.0
C421.1B3.3 GB21 tok/s1592K ctx
dense
EXAONE 4.0 1.2B
C421.2B3.4 GB23 tok/s1453K ctx
dense
Llama 3.2 1B Instruct Q8 0
C421B3.4 GB19 tok/s1731K ctx
dense
TinyLlama 1.1B Chat v0.3
C421.1B3.3 GB21 tok/s1592K ctx
dense
C41
1.1B3.3 GB21 tok/s1592K ctx
dense
C41
0.5B3.0 GB10 tok/s131K ctx
dense
Llama 3.2 1B Instruct
C411B3.2 GB19 tok/s1760K ctx
dense
gemma 3 1b it
C411B3.2 GB19 tok/s1760K ctx
dense
TinyLlama 1.1B Chat v1.0 imatrix
C411.1B3.3 GB21 tok/s1592K ctx
dense
Falcon3 1B Instruct abliterated
C411B3.2 GB19 tok/s1760K ctx
dense
F
F Tier -- Can't Run (127 models)
Too heavy for this hardware.
internlm2 5 20b chat
D4020B17.0 GB29 tok/s9K ctx
dense
internlm2 limarp chat 20b
D4020B17.0 GB29 tok/s9K ctx
dense
internlm2 math plus 20b i1
D4020B17.0 GB29 tok/s9K ctx
dense
Codestral 21B Pruned i1
D3921B17.8 GB24 tok/s4K ctx
dense
Codestral 22B v0.1
D3922B18.5 GB21 tok/s4K ctx
dense
Codestral 22B v0.1
D3922B18.5 GB21 tok/s4K ctx
dense
SmolVLM 500M Instruct
D390.5B3.0 GB10 tok/s2094K ctx
dense
Codestral 22B v0.1
D3922B18.5 GB21 tok/s4K ctx
dense
Codestral 22B v0.1 IMat
D3922B18.5 GB21 tok/s4K ctx
dense
Codestral 22B v0.1 i1
D3922B18.5 GB21 tok/s4K ctx
dense
embeddinggemma 300M
D370.3B2.8 GB6 tok/s2121K ctx
dense
granite embedding 107m multilingual
D340.11B2.7 GB2 tok/s2150K ctx
dense
Falcon H1 Tiny 90M Instruct
D340.09B2.7 GB2 tok/s2151K ctx
dense
F0
70B50.1 GB2 tok/s4K ctx
dense
F0
70B50.1 GB2 tok/s4K ctx
dense
F0
32B25.9 GB6 tok/s4K ctx
dense
F0
72B51.3 GB2 tok/s4K ctx
dense
F0
32B25.9 GB6 tok/s4K ctx
dense
F0
32B25.9 GB6 tok/s4K ctx
dense
F0
235B148.7 GB2 tok/s4K ctx
moe
F0
30.5B22.6 GB24 tok/s4K ctx
moe
F0
30.5B22.6 GB24 tok/s4K ctx
moe
F0
80B52.8 GB6 tok/s4K ctx
moe
F0
480B298.2 GB2 tok/s4K ctx
moe
F0
30B22.3 GB35 tok/s4K ctx
moe
F0
27B22.1 GB11 tok/s4K ctx
dense
F0
35B25.3 GB20 tok/s4K ctx
moe
F0
35B28.0 GB15 tok/s4K ctx
+1moe
F0
27B19.9 GB9 tok/s4K ctx
+1dense
F0
122B79.4 GB4 tok/s4K ctx
moe
F0
397B247.5 GB2 tok/s4K ctx
moe
F0
27B30.2 GB4 tok/s4K ctx
dense
F0
27B30.2 GB4 tok/s4K ctx
dense
F0
24B19.6 GB17 tok/s4K ctx
dense
F0
24B19.6 GB17 tok/s4K ctx
vision
F0
47B33.1 GB6 tok/s4K ctx
moe
F0
141B91.9 GB2 tok/s4K ctx
moe
F0
32B25.9 GB6 tok/s4K ctx
dense
F0
30B24.7 GB7 tok/s4K ctx
dense
F0
35B26.3 GB5 tok/s4K ctx
dense
F0
40B33.1 GB3 tok/s4K ctx
dense
F0
13B22.6 GB26 tok/s7K ctx
dense
F0
15B27.9 GB13 tok/s4K ctx
dense
F0
13B22.6 GB26 tok/s7K ctx
dense
F0
30B47.5 GB3 tok/s4K ctx
dense
F0
24B19.6 GB15 tok/s4K ctx
dense
F0
24B19.6 GB15 tok/s4K ctx
dense
F0
123B82.9 GB2 tok/s4K ctx
dense
F0
119B80.5 GB4 tok/s4K ctx
moe
F0
119B83.9 GB3 tok/s4K ctx
moe
F0
675B419.6 GB2 tok/s4K ctx
+1moe
F0
24B19.6 GB17 tok/s4K ctx
dense
F0
744B475.4 GB2 tok/s4K ctx
moe
F0
754B481.5 GB2 tok/s4K ctx
moe
F0
1000B619.9 GB2 tok/s4K ctx
moe
F0
1000B619.9 GB2 tok/s4K ctx
+1moe
F0
48B32.7 GB2 tok/s4K ctx
linear
F0
9B20.2 GB50 tok/s10K ctx
dense
F0
13B24.1 GB20 tok/s5K ctx
dense
F0
13B24.1 GB20 tok/s5K ctx
dense
F0
13B21.6 GB25 tok/s7K ctx
dense
F0
12B23.3 GB19 tok/s4K ctx
dense
F0
20B35.2 GB6 tok/s4K ctx
dense
F0
34B26.9 GB5 tok/s4K ctx
dense
F0
671B471.4 GB2 tok/s4K ctx
moe
F0
671B471.4 GB2 tok/s4K ctx
moe
F0
1600B866.4 GB2 tok/s4K ctx
moe
F0
284B161.8 GB2 tok/s4K ctx
moe
F0
671B471.4 GB2 tok/s4K ctx
moe
F0
236B205.1 GB2 tok/s4K ctx
moe
F0
236B205.1 GB2 tok/s4K ctx
moe
F0
67B49.2 GB2 tok/s4K ctx
dense
F0
109B71.9 GB3 tok/s4K ctx
moe
F0
400B249.4 GB2 tok/s4K ctx
moe
F0
104B69.4 GB2 tok/s4K ctx
dense
F0
111B74.1 GB2 tok/s4K ctx
dense
F0
34B26.9 GB5 tok/s4K ctx
dense
F0
124B83.5 GB2 tok/s4K ctx
dense
F0
72B51.3 GB2 tok/s4K ctx
dense
F0
72B51.3 GB2 tok/s4K ctx
dense
F0
70B50.1 GB2 tok/s4K ctx
dense
F0
13B22.6 GB26 tok/s4K ctx
dense
F0
13B22.6 GB26 tok/s4K ctx
dense
F0
32B25.9 GB6 tok/s4K ctx
dense
F0
30B23.2 GB8 tok/s4K ctx
dense
F0
32B25.9 GB6 tok/s4K ctx
dense
F0
24B19.6 GB17 tok/s4K ctx
dense
F0
405B257.2 GB2 tok/s4K ctx
dense
F0
70B50.1 GB2 tok/s4K ctx
dense
F0
34B26.9 GB5 tok/s4K ctx
dense
F0
117B78.8 GB2 tok/s4K ctx
dense
F0
32B24.5 GB7 tok/s4K ctx
dense
F0
24B19.6 GB17 tok/s4K ctx
dense
F0
25.2B21.5 GB35 tok/s4K ctx
moe
F0
30.7B35.9 GB2 tok/s4K ctx
dense
F0
671B412.3 GB2 tok/s4K ctx
moe
F0
30B23.7 GB30 tok/s4K ctx
moe
MiniMax M2.7
F0230B146.6 GB2 tok/s4K ctx
moe
Solar Open 100B
F0100B75.2 GB2 tok/s4K ctx
dense
Baichuan M2 32B Q4 K M
F032B25.8 GB6 tok/s4K ctx
dense
baichuan inc Baichuan M2 32B
F032B25.8 GB6 tok/s4K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
F024B20.0 GB15 tok/s4K ctx
dense
cognitivecomputations Dolphin3.0 R1 Mistral 24B
F024B20.0 GB15 tok/s4K ctx
dense
Qwen3 48B A4B Savant Commander Distill 12X Closed Open Heretic Uncensored
F048B37.4 GB2 tok/s4K ctx
dense
EXAONE 4.0 32B
F032B25.8 GB6 tok/s4K ctx
dense
K EXAONE 236B A23B
F0236B174.1 GB2 tok/s4K ctx
dense
Qwen3.5 35B A3B
F035B28.0 GB4 tok/s4K ctx
dense
gemma 3 27b it
F027B22.1 GB10 tok/s4K ctx
dense
Llama 3.3 70B Instruct
F070B53.4 GB2 tok/s4K ctx
dense
Mistral Small 24B Instruct 2501
F024B20.0 GB15 tok/s4K ctx
dense
mistral small 3.1 24b instruct 2503 hf
F024B20.0 GB15 tok/s4K ctx
dense
aya expanse 32b heretic MPOA i1
F032B25.8 GB6 tok/s4K ctx
dense
Baichuan M3 235B
F0235B173.4 GB2 tok/s4K ctx
dense
Baichuan M3 235B i1
F0235B173.4 GB2 tok/s4K ctx
dense
BaichuanMed OCR 72B i1
F072B54.9 GB2 tok/s4K ctx
dense
Dolphin Mistral GLM 4.7 Flash 24B Venice Edition Thinking Uncensored i1
F024B20.0 GB15 tok/s4K ctx
dense
Solar Open 100B i1
F0100B75.2 GB2 tok/s4K ctx
dense
Solar Open 69B REAP i1
F069B52.7 GB2 tok/s4K ctx
dense
Hermes 4.3 36B
F036B28.7 GB4 tok/s4K ctx
dense
stabilityai japanese stablelm instruct beta 70b
F070B53.4 GB2 tok/s4K ctx
dense
F0
27B22.1 GB10 tok/s4K ctx
dense
F0
24B20.0 GB15 tok/s4K ctx
dense
F0
122B76.6 GB2 tok/s4K ctx
dense
F0
27B22.1 GB10 tok/s4K ctx
dense
F0
35B28.0 GB4 tok/s4K ctx
dense
F0
397B291.2 GB2 tok/s4K ctx
dense
cognitivecomputations Dolphin Mistral 24B Venice Edition
F024B20.0 GB15 tok/s4K ctx
dense
Compare with...
See How Other Hardware Stacks Up
Check tier lists for similar GPUs and Apple Silicon configs
