Sign in to MiniCrafts

Sign in to continue.

Continue with Google

Privacy · Terms

AI Benchmarks

Coding leaderboard

LiveBench 2.0 · Coding · score / 100

Synced 2026-11-18

#Model / configuration7-day changeResults
1
91.9
2
91.0
3
89.7
4
89.2
5
87.5
6
87.1
7
85.7
8
84.4
9
84.2
10
83.7
11
82.1
12
80.4
13
79.4
14
79.2
15
78.5
16
77.6
17
77.2
18
76.7
19
74.9
20
73.3
21
73.1
22
70.9
23
70.8
24
69.8
25
69.3
26
68.8
27
67.8
28
67.6
29
64.5
30
60.9
31
55.1
32
54.1
33
53.1
34
51.5
35
48.7
36
44.4
37
34.8
38
20.3
One source and evaluation setup per ranking.

Model details

Compare models

Sources & freshness

LiveBench reports task-based results. Arena reports human preferences. Each board stays within one source and cohort; their scores are never mixed.

Two timestamps, one clear status

Production records will show when a source published a result and when MiniCrafts last synced it. Delayed feeds retain their last successful snapshot with a stale-data label.

How rankings should work

Compare the task you care about, with the source and evaluation setup visible.

A separate ranking for each benchmark

Keep native units, versions and task settings. An arena rating, pass@1 and resolved-task percentage have different meanings.

Agentic, Design and Coding stay distinct

Agentic uses LiveBench 2.0 Long Agentic tasks. Coding and Design use their own suites in the same release. Model settings are retained as published.

Evidence belongs beside the score

Show source-reported sample counts and uncertainty when available. Missing values stay empty; untested models stay unranked.

Match image and video settings

Separate generation from editing, and text-to-video from image-to-video. Arena overall cohorts can include different model defaults. Check the source before comparing audio, resolution, length or quality tiers.

Changes need a stable reference

Only compare trend points within the same benchmark version and setup. A new evaluation version starts a new series.

Share ranking image

View full image