HELM Arabic - Mbzuai Human Translated Arabic Mmlu - Abstract Algebra Leaderboard · The Aggregate

HELM Arabic - Mbzuai Human Translated Arabic Mmlu - Abstract Algebra: leaderboard

Metric: Exact Match (%). Source: crfm.stanford.edu. 41 models tracked.

Top models

#	Model	Score
1	Claude Opus 4.7	82
2	Claude Sonnet 4.6	66
3	Gemma 4 31B (IT)	64
4	GPT-5.4 (2026-03-05)	60
5	Claude Haiku 4.5 (20251001)	59
6	Qwen 3.5 397B A17B	58
7	Gemini 2.5 Flash (Thinking)	57
8	Qwen 2.5 72B Instruct	55
9	Llama 4 Maverick Instruct FP8	54
10	Qwen 3 Next 80B A3B Instruct	53
11	GPT-4.1 Mini	52
12	GPT-4.1 (2025-04-14)	48
13	Gemini 2.5 Flash Lite (Thinking)	47
14	DeepSeek V3.1	45
15	Mistral Large 3	44

Interactive version: theaggregate.ai/benchmark?slug=helm-arabic-mbzuai-human-translated-arabic-mmlu-abstract-algebra · How It Works · Data refreshed daily, snapshot 2026-09-19.