Code Arena🏆Overall

View overall rankings across AI models on agentic coding tasks involving multi-step reasoning and tool use.

Apr 1, 2026
224,709 votes
59 models
Rank Spread
6
515
gpt-5.4-high (codex-harness)
OpenAI · Proprietary
1457+17/-17
1,495N/AN/A
16
719
gpt-5.4-medium (codex-harness)
OpenAI · Proprietary
1427+16/-16
1,579N/AN/A
18
1528
gpt-5.3-codex (codex-harness)
OpenAI · Proprietary
1407+12/-12
2,974$1.75 / $14400K
19
1530
gpt-5.2
OpenAI · Proprietary
1403+17/-17
1,460$1.75 / $14400K
21
1730
gpt-5-medium
OpenAI · Proprietary
1392+13/-13
3,753$1.25 / $10400K
24
1730
gpt-5.1-medium
OpenAI · Proprietary
1390+9/-9
6,124$1.25 / $10400K
29
1732
gpt-5.4-mini-high
OpenAI · Proprietary
1385+18/-18
1,198$0.75 / $4.50400K
35
3340
gpt-5.1
OpenAI · Proprietary
1339+7/-7
12,868$1.25 / $10400K
37
3440
gpt-5.2-codex
OpenAI · Proprietary
1335+8/-8
7,956$1.75 / $14400K
39
3440
gpt-5.1-codex
OpenAI · Proprietary
1328+9/-9
6,225$1.25 / $10400K
49
4653
gpt-5.1-codex-mini
OpenAI · Proprietary
1238+17/-17
1,443$0.25 / $2400K

Remove Style Control Leaderboard Plots

Confidence Intervals on Model Strength (via Bootstrapping)

Battle Count for Each Combination of Models (without Ties)

Average Win Rate Against All Other Models (Uniform Sampling and No Ties)

Fraction of Model A Wins for All Non-tied A vs. B Battles