Skip to content
LLM Pulse

Leaderboard

Artificial Analysis Coding Agent Index

17 models scored · metric: average pass@1. Prices are the cheapest listed across all serving providers. Pts/$ = score ÷ blended price (3×input + output) / 4 — how much score each dollar buys.

Best value · top 5 by pts per dollar
  1. 1.GPT-5.6 Luna332
  2. 2.DeepSeek V4 Pro92.1
  3. 3.Kimi K2.649.3
  4. 4.GPT-5.6 Terra34.4
  5. 5.GLM-5.124.5
#ModelScoreBest in /MBest out /MPts / $Context
🥇GPT-5.6 Solopenai
80
$2.50$1514.21.05M
🥈GPT-5.6 Terraopenai
77.4
$1$634.41.05M
🥉GPT-5.6 Lunaopenai
74.6
$0.1$0.63321.05M
4Claude Opus 4.7anthropic
66.6
$5$256.71M
5GPT-5.5openai
65.3
$4.55$27.276.41.05M
6Claude Opus 4.7anthropic
61.2
$5$256.11M
7GPT-5.5openai
60.4
$4.55$27.275.91.05M
8Claude Opus 4.7anthropic
59.9
$5$256.01M
9GPT-5.5openai
57.8
$4.55$27.275.71.05M
10GPT-5.4openai
53.6
$2.20$1410.41.05M
11GLM-5.1zhipuai
52.7
$1.40$4.4024.5200K
12GPT-5.4openai
52.2
$2.20$1410.11.05M
13Claude Opus 4.6anthropic
51.3
$5$255.11M
14Kimi K2.6moonshotai
50.5
$0.5$2.6049.3262K
15DeepSeek V4 Prodeepseek
50.1
$0.435$0.8792.11M
16Claude Sonnet 4.6anthropic
49.4
$3$158.21M
17Gemini 3.1 Pro Previewgoogle
43
$2$129.61.05M

Cheapest scorer: GPT-5.6 Luna at $0.1 input /M.