TIDE (Evolving Documents, Gold Context): leaderboard
Metric: Macro-averaged strict accuracy (%; mean over the eight task types; TIDE, 3,050 expert-verified questions over 644 Bangladesh customs instruments (1969-2025); free-form answers are judged by a council of Claude Opus 4.6, Gemini 3.5 Flash and GPT-5.1 by majority vote with a hard date gate, deterministic tasks by exact match; strict accuracy without the partial-rubric credit; thinking enabled, temperature 0; gold context: the governing instrument text is supplied in the prompt). Source: arxiv.org. Saturation forecast: Around July 2027. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 (Thinking) | 68.49 |
| 2 | Claude Sonnet 4.5 (Thinking) | 64.39 |
| 3 | Kimi K2.5 (Thinking) | 63.27 |
| 4 | Gemini 3.5 Flash | 62.45 |
| 5 | Qwen 3.6 35B A3B | 61.13 |
| 6 | Gemma 4 31B (Reasoning) | 60.05 |
| 7 | GPT-OSS-120B | 57.11 |
| 8 | GLM-4.6 | 50.38 |
Interactive version: theaggregate.ai/benchmark?slug=tide-evolving-documents-gold-context · How It Works · Data refreshed daily, snapshot 2026-09-29.