Skip to content
LLM Pulse

Leaderboard

FrontierMath

11 models scored · metric: accuracy. 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 Luna349
  2. 2.GPT-5.6 Terra37.7
  3. 3.GPT-5.6 Sol15.8
  4. 4.GPT-5.49.2
  5. 5.GPT-5.45.3
#ModelScoreBest in /MBest out /MPts / $Context
🥇GPT-5.6 Solopenai
89
$2.50$1515.81.05M
🥈GPT-5.6 Terraopenai
84.9
$1$637.71.05M
🥉GPT-5.6 Lunaopenai
78.6
$0.1$0.63491.05M
4GPT-5.5 Proopenai
52.4
$27.27$163.640.91.05M
5GPT-5.5openai
51.7
$4.55$27.275.11.05M
6GPT-5.4 Proopenai
50
$27$1600.81.05M
7GPT-5.4openai
47.6
$2.20$149.21.05M
8GPT-5.5 Proopenai
39.6
$27.27$163.640.61.05M
9GPT-5.4 Proopenai
38
$27$1600.61.05M
10GPT-5.5openai
35.4
$4.55$27.273.51.05M
11GPT-5.4openai
27.1
$2.20$145.31.05M

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