Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

HELM SeaHELM - XNLI (th) — leaderboard

Metric: EM. Source: crfm.stanford.edu. 21 models tracked.

Top models

#ModelScore
1Llama 3.1 405B Instruct70
2GPT-4o (2024-11-20)69.4
3Claude 3.7 Sonnet (20250219)66.3
4Qwen 2 72B Instruct63.5
5Gemini 2.0 Flash63
6Llama 3.3 70B Instruct60.4
7Llama 3.1 70B Instruct58.9
8Gemini 2.0 Flash Lite57.2
9Qwen 2.5 72B Instruct56.9
10Llama 3.2 90B Vision Instruct56.1
11GPT-4o Mini (2024-07-18)56
12Claude 3.5 Haiku (20241022)55.1
13Llama 4 Maverick Instruct FP850.7
14DeepSeek V350.5
15DeepSeek R1 Distill Llama 8B33.3

Interactive version: theaggregate.ai/benchmark?slug=helm-seahelm-xnli-th · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.