Dataset Viewer
Auto-converted to Parquet Duplicate
model
stringlengths
18
35
family
stringclasses
6 values
cpu_tps
float64
2.6
59.7
phone_tps
unknown
size_mb
float64
0
4.47k
chat_format
stringclasses
3 values
speed_per_mb
float64
0
0.11
SmolLM2-135M-Instruct-mobile
SmolLM2
59.7
46
0
llama-3
0
SmolLM2-360M-Instruct-mobile
SmolLM2
29.1
""
258.1
chatml
0.113
MiniCPM-V-4.6-mobile
MiniCPM
23.4
""
504.6
chatml
0.046
SmolLM2-1.7B-Instruct-mobile
SmolLM2
18.7
""
1,006.7
chatml
0.019
MiniCPM5-1B-multilingual-mobile
MiniCPM
18.4
""
656.2
null
0.028
MiniCPM5-1B-mobile
MiniCPM
18.1
""
656.2
null
0.028
TinyLlama-1.1B-Chat-mobile-int4
Llama-3.2
17.6
""
637.8
chatml
0.028
Qwen2.5-0.5B-Instruct-Server-mobile
Qwen2.5
16.7
""
482.3
chatml
0.035
Qwen2.5-0.5B-Chinese-mobile
Qwen2.5
16.4
""
400.6
chatml
0.041
Qwen2.5-Math-1.5B-mobile
Qwen2.5
15.7
""
940.4
chatml
0.017
Gemma-2B-Arabic-mobile
Gemma-2
14.8
null
1,554.7
gemma
0.01
Qwen2.5-0.5B-Instruct-mobile-int4
Qwen2.5
12.5
23.2
0
chatml
0
Llama-3.2-1B-Instruct-Q4-mobile
Llama-3.2
11.3
5.4
0
chatml
0
TinyLlama-1.1B-Chat-Q5-mobile
Llama-3.2
10.9
""
746.7
chatml
0.015
SmolLM2-1.7B-Instruct-Q5-mobile
SmolLM2
10.8
""
1,168.7
chatml
0.009
Llama-3.2-3B-FunctionCall-mobile
Llama-3.2
9.1
""
1,925.8
llama-3
0.005
Llama-3.2-1B-FunctionCall-mobile
Llama-3.2
8.9
null
1,925.8
chatml
0.005
Qwen2.5-1.5B-Instruct-Q8-mobile
Qwen2.5
8.7
""
1,570.3
chatml
0.006
Phi-3.5-mini-Instruct-mobile
Phi-3.5
8.6
""
2,282.4
chatml
0.004
Gemma-2-2B-IT-mobile
Gemma-2
8.2
""
1,629.4
gemma
0.005
Qwen2.5-0.5B-Coder-mobile
Qwen2.5
8
""
379.4
chatml
0.021
Qwen2.5-Coder-1.5B-mobile
Qwen2.5
7.8
""
940.4
chatml
0.008
Qwen2.5-1.5B-Instruct-mobile-int4
Qwen2.5
7.7
""
1,065.6
chatml
0.007
Phi-3.5-mini-instruct-Q5-mobile
Phi-3.5
7
""
2,684.9
chatml
0.003
Gemma-2-2B-IT-Q5-mobile
Gemma-2
5.7
""
1,834.2
gemma
0.003
Llama-3.2-1B-Instruct-Q6-mobile
Llama-3.2
5
""
974.5
llama-3
0.005
Llama-3.2-1B-Instruct-mobile
Llama-3.2
4.9
""
1,259.9
llama-3
0.004
Llama-3.2-3B-Instruct-mobile
Llama-3.2
4.8
""
1,925.8
llama-3
0.002
Qwen2.5-1.5B-Instruct-Q5-mobile
Qwen2.5
4.4
""
1,072.9
chatml
0.004
Qwen2.5-Coder-7B-mobile
Qwen2.5
3
""
4,466.1
chatml
0.001
Llama-3.2-3B-Instruct-Q5-mobile
Llama-3.2
2.6
""
2,214.6
llama-3
0.001

Detailed Comparison

Real verified data for all 31 working models. Includes speed-per-MB efficiency metric.

๐Ÿš€ dispatchAI

Downloads last month
-