Leaderboard
GDPval-AA
9 models scored · metric: Elo. 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.Gemini 3.5 Flash Lite1,341
- 2.Grok 4.3960
- 3.Gemini 3.6 Flash947
- 4.Gemini 3.5 Flash491
- 5.Kimi K3318
| # | Model | Score | Best in /M | Best out /M | Pts / $ | Context |
|---|---|---|---|---|---|---|
| 🥇 | Claude Fable 5anthropic | 1932 | $10 | $50 | 96.6 | 1M |
| 🥈 | Claude Opus 5anthropic | 1861 | $5 | $25 | 186 | 1M |
| 🥉 | Kimi K3moonshotai | 1668 | $2.50 | $13.50 | 318 | 1.05M |
| 4 | Gemini 3.5 Flashgoogle | 1656 | $1.50 | $9 | 491 | 1.05M |
| 5 | Grok 4.3xai | 1500 | $1.25 | $2.50 | 960 | 1M |
| 6 | Gemini 3.6 Flashgoogle | 1421 | $0.75 | $3.75 | 947 | 1.05M |
| 7 | Gemini 3.1 Pro Previewgoogle | 1314 | $2 | $12 | 292 | 1.05M |
| 8 | Gemini 3.5 Flash Litegoogle | 1140 | $0.3 | $2.50 | 1,341 | 1.05M |
| 9 | North Mini Codecohere | 14 | — | — | — | 256K |
Cheapest scorer: Gemini 3.5 Flash Lite at $0.3 input /M.