TimeLitmus - Finance Natural Prediction: leaderboard

Metric: Natural accuracy (%; 380 Finance records of FNSPID market events with four expected-response classes (strong or weak, up or down); ten LLMs with deterministic decoding where available, invalid or unparsable outputs scored incorrect). Source: arxiv.org. Saturation forecast: Around 2033. 10 models tracked.

Top models

#ModelScore
1GLM-551.3
2DeepSeek V4 Flash48.7
3MiniMax-M347.1
4GPT-5.444.7
5DeepSeek R1 052844.5
6Gemini 3.5 Flash42.1
7Qwen Plus41.1
8Qwen 3.5 4B40.8
9Qwen 3.5 9B40.3
10Claude Sonnet 4.637.1

Interactive version: theaggregate.ai/benchmark?slug=timelitmus-finance-natural-prediction · How It Works · Data refreshed daily, snapshot 2026-09-26.