AI Model Leaderboards

AI Gateway's model leaderboard for the period from June 21, 2026 to September 18, 2026 shows Jev leading in reach at 15.6% of teams and in preference at 13.3% of teams using it as their primary model by token volume. Claude Haiku 4.5 ranked second in reach at 14.6%, followed by GPT 5.6 Luna at 13.0% and Claude Sonnet 5 at 12.2%, while the top 10 models accounted for 58.4% of requests and 67.8% of spend, leaving 41.6% and 32.2% respectively to other models.

AI model leaderboards See top models, labs, apps, and providers on AI Gateway with token, request, and cost trends. Copy link to heading token-volume Token Volume Share of AI Gateway token volume across the top models. Jun 21, 2026Sep 18, 2026 10Other 16.5% Copy link to heading open-vs-closed Open vs. Closed Token Volume Share of AI Gateway token volume on models that publish their weights for download, versus everything else. Jun 21, 2026Sep 18, 2026 Copy link to heading requests Requests Share of AI Gateway requests across the top models. Jun 21, 2026Sep 18, 2026 10Other 41.6% Copy link to heading spend Spend Share of AI Gateway spend across the top models. Jun 21, 2026Sep 18, 2026 10Other 32.2% Copy link to heading reach Reach Share of AI Gateway teams using each model. Teams use several models, so these shares do not sum to 100%. Jun 21, 2026Sep 18, 2026 1Jev 15.6% 2Claude Haiku 4.5 14.6% 3GPT 5.6 Luna 13.0% 4Claude Sonnet 5 12.2% 5text-embedding-3-small 8.7% 6Claude Sonnet 4.6 8.3% 7Gemini 2.5 Flash 6.5% 8Gemini 2.5 Flash Lite 6.5% 9GPT 5.6 Sol 6.2% 10Gemini 3.8 Flash 6.2% Copy link to heading preference Preference Share of AI Gateway teams with each model as their primary model, by token volume. Jun 21, 2026Sep 18, 2026 1Jev 13.3% 2GPT 5.6 Luna 6.9% 3Claude Sonnet 5 6.9% 4Claude Haiku 4.5 6.7% 5Claude Sonnet 4.6 4.2% 6Gemini 2.5 Flash 3.2% 7GPT-4o mini 2.9% 8Gemini 2.5 Flash Lite 2.9% 9GPT-5 mini 2.7% 10Other 50.5%