TIGER-Lab: AI model benchmarks

AI models from TIGER-Lab. 4 models tracked, 4 with a unified ELO rating. Current best: AceCoder-Qwen2.5-7B-Ins-Rule (ELO 1537, rank #906 overall).

Rated models

RankModelProviderUnified ELOBenchmarks
906AceCoder-Qwen2.5-7B-Ins-RuleTIGER-Lab1537 ± 5911
1032AceCodeRM-7BTIGER-Lab1512 ± 5711
1146AceCoder-Qwen2.5-Coder-7B-Ins-RuleTIGER-Lab1495 ± 5811
1892AceCoder-Qwen2.5-Coder-7B-Base-RuleTIGER-Lab1402 ± 6111

Interactive version: theaggregate.ai/company?slug=tiger-lab · How It Works · Data refreshed daily, snapshot 2026-09-21.