Counts across the 1560 projects in both rankings, from the same data as the lists.
Data from 24 Oct 2026, refreshed twice a day. The open data API has every project behind these counts.
Projects1560
Categories33
Stars, all projects30.9m
Median score66
Method changes4
Projects per category
The biggest categories are Skills, plugins and prompts on Best of Vibe Coding (308 projects) and ML frameworks and libraries on Best of Open-Source AI (120 projects).
83% of all 1560 projects use a permissive license such as MIT or Apache-2.0. 173 have custom terms or no license, so read those before you use one at work.
Best of Vibe CodingNumbers as a table
License families, Best of Vibe Coding
License
Projects
Share
Permissive (MIT, Apache, BSD)
665
87%
Copyleft (GPL, AGPL)
36
5%
Custom or none
63
8%
Best of Open-Source AINumbers as a table
License families, Best of Open-Source AI
License
Projects
Share
Permissive (MIT, Apache, BSD)
631
79%
Copyleft (GPL, AGPL)
55
7%
Custom or none
110
14%
GPU needs
On Best of Open-Source AI, 268 of 796 projects need no GPU and 319 run faster with one; 88 require a GPU.
Best of Vibe CodingNumbers as a table
GPU needs, from each README, Best of Vibe Coding
GPU
Projects
Share
No GPU needed
658
86%
A GPU helps, not required
34
4%
Needs a GPU
2
<1%
README does not say
70
9%
Best of Open-Source AINumbers as a table
GPU needs, from each README, Best of Open-Source AI
GPU
Projects
Share
No GPU needed
268
34%
A GPU helps, not required
319
40%
Needs a GPU
88
11%
README does not say
121
15%
Docker, arm64 and live demos
Docker files or an image ship with 25% of Best of Vibe Coding and 48% of Best of Open-Source AI. You can try 326 projects in a live demo before you install anything.
Best of Vibe CodingNumbers as a table
Ways to run it, Best of Vibe Coding
Has
Projects
Share
Docker
192
25%
arm64 or Apple Silicon
214
28%
Live demo
111
15%
Best of Open-Source AINumbers as a table
Ways to run it, Best of Open-Source AI
Has
Projects
Share
Docker
385
48%
arm64 or Apple Silicon
259
33%
Live demo
215
27%
Languages
The most common main language is TypeScript on Best of Vibe Coding (323) and Python on Best of Open-Source AI (459).
Best of Vibe CodingNumbers as a table
Main language on GitHub, Best of Vibe Coding
Language
Projects
Share
TypeScript
323
42%
Python
172
23%
JavaScript
54
7%
Rust
53
7%
Go
47
6%
Shell
22
3%
Other languages
74
10%
No main language
19
2%
Best of Open-Source AINumbers as a table
Main language on GitHub, Best of Open-Source AI
Language
Projects
Share
Python
459
58%
TypeScript
111
14%
Jupyter Notebook
66
8%
C++
38
5%
Rust
37
5%
Go
27
3%
Other languages
54
7%
No main language
4
<1%
Last commit
1133 of 1560 projects (73%) had a commit in the last 30 days. 4 have gone more than a year without one.
Best of Vibe CodingNumbers as a table
Last commit to the default branch, Best of Vibe Coding
Last commit
Projects
Share
In the last 30 days
568
74%
31 to 90 days ago
89
12%
91 days to a year ago
105
14%
More than a year ago
2
<1%
Best of Open-Source AINumbers as a table
Last commit to the default branch, Best of Open-Source AI
Last commit
Projects
Share
In the last 30 days
565
71%
31 to 90 days ago
85
11%
91 days to a year ago
144
18%
More than a year ago
2
<1%
Stars
The median project has 3.7k stars on Best of Vibe Coding and 14.2k stars on Best of Open-Source AI. OpenClaw has the most: 392.9k.
Best of Vibe CodingNumbers as a table
GitHub stars, Best of Vibe Coding
Stars
Projects
Share
100k or more
22
3%
50k to 100k
26
3%
20k to 50k
84
11%
5k to 20k
195
26%
1k to 5k
383
50%
Under 1k
54
7%
Best of Open-Source AINumbers as a table
GitHub stars, Best of Open-Source AI
Stars
Projects
Share
100k or more
23
3%
50k to 100k
65
8%
20k to 50k
202
25%
5k to 20k
475
60%
1k to 5k
30
4%
Under 1k
1
<1%
Scores
The median score is 67 on Best of Vibe Coding and 64 on Best of Open-Source AI, and 322 projects score 80 or more out of 100.
Best of Vibe CodingNumbers as a table
Scores out of 100, Best of Vibe Coding
Score
Projects
Share
90 to 100
46
6%
80 to 89
116
15%
70 to 79
181
24%
60 to 69
164
21%
50 to 59
102
13%
40 to 49
69
9%
30 to 39
41
5%
Under 30
45
6%
Best of Open-Source AINumbers as a table
Scores out of 100, Best of Open-Source AI
Score
Projects
Share
90 to 100
33
4%
80 to 89
127
16%
70 to 79
146
18%
60 to 69
162
20%
50 to 59
113
14%
40 to 49
93
12%
30 to 39
52
7%
Under 30
70
9%
Method changes
4 dated changes to how we rank, newest first. The methodology page keeps the same log.
2026-10-14: the star-growth anomaly penalty is dropped, since a star chart cannot tell a big launch from bought stars
2026-10-13: Maintenance weighs 15% (was 10%) and Momentum 10% (was 15%), so steady upkeep counts more than a burst of stars
2026-10-12: penalties we do not check yet are marked coming; only "no license" has ever been applied
2026-10-09: Docker alone earns no points, medals mark places 1–3 instead of score bands