UniQL - Hive: leaderboard

Metric: Execution accuracy (%) on the Hive dialect of UniQL (1,534 BIRD development-set questions with human-verified SQL in 16 aligned dialects): given the target dialect, schema and question, the model writes one query whose execution result must match the gold result; inference-only, one run; higher is better. Source: arxiv.org. Saturation forecast: Around February 2028. 13 models tracked.

Top models

#ModelScore
1Claude Sonnet 4.559.58
2Gemini 2.5 Pro59.32
3GPT-5.1 Codex55.48
4GPT-5 Mini54.04
5DeepSeek V4 Flash52.8
6Qwen 3 32B51.96
7Qwen 3 4B46.74
8Qwen 3 8B44.85
9Llama 3 70B Instruct42.63
10Qwen 3 1.7B36.7
11GPT-3.5 Turbo36.18
12Llama 3 8B Instruct22.23

Interactive version: theaggregate.ai/benchmark?slug=uniql-hive · How It Works · Data refreshed daily, snapshot 2026-09-29.