Model
Fit
Speed
Memory
Scores / Run
Runs comfortably
Ornith-1.5-35B-A3B
Q8_0
37.8GB
262k ctx
Runs comfortably
117
tok/s
ⓘ estimated
38
/ 64GB
91
/
89
/
86
Runs comfortably
117 tok/s
ⓘ estimated
38/64GB
Ornith-1.5-35B-A3B
Q6_K
29.2GB
262k ctx
Runs comfortably
151
tok/s
ⓘ estimated
29
/ 64GB
91
/
89
/
86
Runs comfortably
151 tok/s
ⓘ estimated
29/64GB
Ornith-1.5-35B-A3B
Q5_K_M
25.3GB
262k ctx
Runs comfortably
175
tok/s
ⓘ estimated
25
/ 64GB
91
/
89
/
86
Runs comfortably
175 tok/s
ⓘ estimated
25/64GB
Ornith-1.5-35B-A3B
Q4_K_M
21.7GB
262k ctx
Runs comfortably
203
tok/s
ⓘ estimated
22
/ 64GB
91
/
89
/
86
Runs comfortably
203 tok/s
ⓘ estimated
22/64GB
Qwen3.6 35B A3B
Q4_K_M
20.0GB
262k ctx
Runs comfortably
215
tok/s
ⓘ estimated
20
/ 64GB
90
/
91
/
87
Runs comfortably
215 tok/s
ⓘ estimated
20/64GB
Runs comfortably
116 tok/s
ⓘ estimated
37/64GB
Pokee-Isaac 28B
Q4_K_M
16.0GB
10000k ctx
Runs comfortably
23
tok/s
ⓘ estimated
16
/ 64GB
90
/
88
/
88
Runs comfortably
23 tok/s
ⓘ estimated
16/64GB
Runs comfortably
13 tok/s
ⓘ estimated
28/64GB
Runs comfortably
7 tok/s
ⓘ estimated
56/64GB
15.2GB
262k ctx
Runs comfortably
24
tok/s
ⓘ estimated
15
/ 64GB
88
/
89
/
85
Runs comfortably
24 tok/s
ⓘ estimated
15/64GB
6.2GB
262k ctx
Runs comfortably
59
tok/s
ⓘ estimated
6
/ 64GB
88
/
89
/
85
Runs comfortably
59 tok/s
ⓘ estimated
6/64GB
9.8GB
262k ctx
Runs comfortably
37
tok/s
ⓘ estimated
10
/ 64GB
88
/
89
/
85
Runs comfortably
37 tok/s
ⓘ estimated
10/64GB
17.6GB
262k ctx
Runs comfortably
21
tok/s
ⓘ estimated
18
/ 64GB
88
/
89
/
85
Runs comfortably
21 tok/s
ⓘ estimated
18/64GB
Runs comfortably
23 tok/s
ⓘ estimated
16/64GB
Runs comfortably
20 tok/s
ⓘ estimated
18/64GB
Runs comfortably
11 tok/s
ⓘ estimated
33/64GB
Runs comfortably
20 tok/s
ⓘ estimated
18/64GB
Runs comfortably
174 tok/s
ⓘ estimated
14/64GB
Gemma 4 26B A4B
Q4_K_M
13.0GB
256k ctx
Runs comfortably
188
tok/s
ⓘ estimated
13
/ 64GB
85
/
88
/
84
Runs comfortably
188 tok/s
ⓘ estimated
13/64GB
Runs comfortably
98 tok/s
ⓘ estimated
25/64GB
Runs comfortably
49 tok/s
ⓘ estimated
50/64GB
Runs comfortably
22 tok/s
ⓘ estimated
17/64GB
Runs comfortably
13 tok/s
ⓘ estimated
28/64GB
Runs comfortably
13 tok/s
ⓘ estimated
29/64GB
Runs comfortably
15 tok/s
ⓘ estimated
25/64GB
Runs comfortably
11 tok/s
ⓘ estimated
34/64GB
Runs comfortably
18 tok/s
ⓘ estimated
20/64GB
Nemotron 3.5 Lightning
Q4_K_M
32.0GB
1000k ctx
Runs comfortably
101
tok/s
ⓘ estimated
32
/ 64GB
78
/
76
/
74
Runs comfortably
101 tok/s
ⓘ estimated
32/64GB
Runs comfortably
55 tok/s
ⓘ estimated
7/64GB
Runs comfortably
49 tok/s
ⓘ estimated
8/64GB
Runs comfortably
64 tok/s
ⓘ estimated
6/64GB
Runs comfortably
38 tok/s
ⓘ estimated
10/64GB
Runs comfortably
20 tok/s
ⓘ estimated
18/64GB
Runs comfortably
241 tok/s
ⓘ estimated
16/64GB
Runs comfortably
121 tok/s
ⓘ estimated
31/64GB
Runs comfortably
43 tok/s
ⓘ estimated
9/64GB
Runs comfortably
13 tok/s
ⓘ estimated
29/64GB
Runs comfortably
25 tok/s
ⓘ estimated
15/64GB
Xing4.0-29B-A4B
Q4_K_M
20.1GB
256k ctx
Runs comfortably
133
tok/s
ⓘ estimated
20
/ 64GB
75
/
73
/
72
Runs comfortably
133 tok/s
ⓘ estimated
20/64GB
Llama 3.3 70B Instruct
Q3_K_M
30.0GB
128k ctx
Runs comfortably
12
tok/s
ⓘ estimated
30
/ 64GB
72
/
77
/
80
Runs comfortably
12 tok/s
ⓘ estimated
30/64GB
21.5GB
128k ctx
Runs comfortably
17
tok/s
ⓘ estimated
22
/ 64GB
72
/
77
/
80
Runs comfortably
17 tok/s
ⓘ estimated
22/64GB
Llama 3.3 70B Instruct
Q4_K_M
42.0GB
128k ctx
Runs comfortably
9
tok/s
ⓘ estimated
42
/ 64GB
72
/
77
/
80
Runs comfortably
9 tok/s
ⓘ estimated
42/64GB
Runs comfortably
41 tok/s
ⓘ estimated
9/64GB
Runs comfortably
25 tok/s
ⓘ estimated
15/64GB
Runs comfortably
51 tok/s
ⓘ estimated
7/64GB
Runs comfortably
28 tok/s
ⓘ estimated
13/64GB
Runs comfortably
13 tok/s
ⓘ estimated
28/64GB
Runs comfortably
22 tok/s
ⓘ estimated
17/64GB
Runs comfortably
71 tok/s
ⓘ estimated
5/64GB
Runs comfortably
43 tok/s
ⓘ estimated
9/64GB
Mixtral 8x7B Instruct
Q3_K_M
19.0GB
32k ctx
Runs comfortably
70
tok/s
ⓘ estimated
19
/ 64GB
62
/
65
/
68
Runs comfortably
70 tok/s
ⓘ estimated
19/64GB
Mixtral 8x7B Instruct
Q4_K_M
26.5GB
32k ctx
Runs comfortably
50
tok/s
ⓘ estimated
27
/ 64GB
62
/
65
/
68
Runs comfortably
50 tok/s
ⓘ estimated
27/64GB
Runs comfortably
27 tok/s
ⓘ estimated
14/64GB
Runs comfortably
49 tok/s
ⓘ estimated
8/64GB
Runs comfortably
50 tok/s
ⓘ estimated
7/64GB
Runs comfortably
28 tok/s
ⓘ estimated
13/64GB
Runs comfortably
142 tok/s
ⓘ estimated
3/64GB
Runs comfortably
147 tok/s
ⓘ estimated
3/64GB
Runs comfortably
90 tok/s
ⓘ estimated
4/64GB
16.0GB
128k ctx
Runs comfortably
23
tok/s
ⓘ estimated
16
/ 64GB
52
/
55
/
62
Runs comfortably
23 tok/s
ⓘ estimated
16/64GB
8.0GB
128k ctx
Runs comfortably
46
tok/s
ⓘ estimated
8
/ 64GB
52
/
55
/
62
Runs comfortably
46 tok/s
ⓘ estimated
8/64GB
Llama 3.1 8B Instruct
Q4_K_M
4.5GB
128k ctx
Runs comfortably
82
tok/s
ⓘ estimated
5
/ 64GB
52
/
55
/
62
Runs comfortably
82 tok/s
ⓘ estimated
5/64GB
Runs comfortably
123 tok/s
ⓘ estimated
3/64GB
Runs comfortably
74 tok/s
ⓘ estimated
5/64GB
3.5GB
128k ctx
Runs comfortably
105
tok/s
ⓘ estimated
4
/ 64GB
35
/
38
/
50
Runs comfortably
105 tok/s
ⓘ estimated
4/64GB
Llama 3.2 3B Instruct
Q4_K_M
2.0GB
128k ctx
Runs comfortably
184
tok/s
ⓘ estimated
2
/ 64GB
35
/
38
/
50
Runs comfortably
184 tok/s
ⓘ estimated
2/64GB
Runs comfortably
11 tok/s
ⓘ estimated
33/64GB
Runs comfortably
20 tok/s
ⓘ estimated
18/64GB
Runs tight
Qwen3.8-Flash-Next
EXL3_2BPW
62.7GB
262k ctx
Runs tight
176
tok/s
ⓘ estimated
63
/ 64GB
92
/
92
/
89
Runs tight
176 tok/s
ⓘ estimated
63/64GB
Runs tight
6 tok/s
ⓘ estimated
61/64GB
Runs tight
54 tok/s
ⓘ estimated
60/64GB
Runs tight
56 tok/s
ⓘ estimated
58/64GB
Runs tight
43 tok/s
ⓘ estimated
62/64GB
Coding alternatives
Hosted coding assistants for comparison - per seat
Cursor
Hobby
AI-first code editor. Uses frontier models (GPT-4, Claude). $20/user Business...
$0.0
GitHub Copilot
Individual
AI coding assistant by GitHub. GPT-4o and Claude-based completions. $19/seat ...
$10.0
GitHub Copilot
Business
AI coding assistant by GitHub. GPT-4o and Claude-based completions. $19/seat ...
$19.0
Cursor
Pro
AI-first code editor. Uses frontier models (GPT-4, Claude). $20/user Business...
$20.0
GitHub Copilot
Enterprise
AI coding assistant by GitHub. GPT-4o and Claude-based completions. $19/seat ...
$39.0
Cursor
Business
AI-first code editor. Uses frontier models (GPT-4, Claude). $20/user Business...
$40.0
Per-token monthly figures are estimates at typical usage. Prices verified periodically - always confirm current pricing on the provider's website.
Sign in with GitHub to save this rig and get weekly model updates
Sign in with GitHub