Skip to content

Benchmark

SeaLLMs SeaExam public

Multilingual benchmarks compare model performance across languages. Scores are percentages. Higher is better.

Models measured
57
Scores recorded
57
Sources
1
IQ weight
1.40

Score distribution

60.9%
Mean
63.4%
Median
13.4%
Std dev
43.8%
10th percentile
75.6%
90th percentile
28.3%
Min
28.3%91.2%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.40 for chat models, measured over 57 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 2.5 Pro Preview 03-25· SeaLLMs Leaderboard for SEA Languages data docs91.2%
2DeepSeek V3· SeaLLMs Leaderboard for SEA Languages data docs78.8%
3Claude 3.5 Sonnet· SeaLLMs Leaderboard for SEA Languages data docs78.7%
4Claude Sonnet 3.7· SeaLLMs Leaderboard for SEA Languages data docs78.7%
5gemini-pro-1.5· SeaLLMs Leaderboard for SEA Languages data docs77.6%
6Gemini 2.0 Flash· SeaLLMs Leaderboard for SEA Languages data docs77.5%
7Qwen2.5 72B· SeaLLMs Leaderboard for SEA Languages data docs75.6%
8Qwen2.5 72B Instruct· SeaLLMs Leaderboard for SEA Languages data docs75.2%
9Llama 4 Maverick· SeaLLMs Leaderboard for SEA Languages data docs75%
10GPT 4.1· SeaLLMs Leaderboard for SEA Languages data docs74.8%
11GPT 4o· SeaLLMs Leaderboard for SEA Languages data docs74.6%
12Qwen 2 (72B)· SeaLLMs Leaderboard for SEA Languages data docs74.6%
13Gemini 2.0 Flash Lite· SeaLLMs Leaderboard for SEA Languages data docs74.3%
14Qwen2 72B Instruct· SeaLLMs Leaderboard for SEA Languages data docs72.3%
15Gemma 3 27B Pt· SeaLLMs Leaderboard for SEA Languages data docs71.4%
16Qwen2.5 14B· SeaLLMs Leaderboard for SEA Languages data docs70.5%
17GPT-4.1 mini· SeaLLMs Leaderboard for SEA Languages data docs70%
18Qwen 2.5 14B Instruct· SeaLLMs Leaderboard for SEA Languages data docs69.8%
19Llama 3.1 70B Instruct· SeaLLMs Leaderboard for SEA Languages data docs69.7%
20Gemini 1.5 Flash· SeaLLMs Leaderboard for SEA Languages data docs69.4%
21Gemma 3 27B IT· SeaLLMs Leaderboard for SEA Languages data docs68.4%
22Llama 3 70B Instruct· SeaLLMs Leaderboard for SEA Languages data docs66%
23Llama 3.1 70B· SeaLLMs Leaderboard for SEA Languages data docs65.6%
24Gemma 2 27B IT· SeaLLMs Leaderboard for SEA Languages data docs65.1%
25Meta Llama 3 70b· SeaLLMs Leaderboard for SEA Languages data docs64.9%
26GPT-4o mini· SeaLLMs Leaderboard for SEA Languages data docs64.2%
27Llama 4 Scout· SeaLLMs Leaderboard for SEA Languages data docs63.9%
28Claude Haiku 3.5· SeaLLMs Leaderboard for SEA Languages data docs63.7%
29Qwen2.5 7B· SeaLLMs Leaderboard for SEA Languages data docs63.4%
30Gemma 3 12B IT· SeaLLMs Leaderboard for SEA Languages data docs63.2%
31Qwen2.5 7B Instruct· SeaLLMs Leaderboard for SEA Languages data docs62.6%
32Gemma 2 9B IT· SeaLLMs Leaderboard for SEA Languages data docs62.5%
33gemma2-27b· SeaLLMs Leaderboard for SEA Languages data docs61.7%
34Claude 3 Haiku· SeaLLMs Leaderboard for SEA Languages data docs59.1%
35gemma-2-9b· SeaLLMs Leaderboard for SEA Languages data docs58.9%
36GPT-4.1 nano· SeaLLMs Leaderboard for SEA Languages data docs58.4%
37Qwen2 7B Instruct· SeaLLMs Leaderboard for SEA Languages data docs57.6%
38Qwen2.5 3B Instruct· SeaLLMs Leaderboard for SEA Languages data docs55.5%
39GLM 4 9B 0414· SeaLLMs Leaderboard for SEA Languages data docs54.9%
40GLM-4-9b-chat· SeaLLMs Leaderboard for SEA Languages data docs53.9%
41Llama 3.1 8B Instruct· SeaLLMs Leaderboard for SEA Languages data docs52.7%
42Llama 3.1 8B· SeaLLMs Leaderboard for SEA Languages data docs50.3%
43Gemma 3 4B IT· SeaLLMs Leaderboard for SEA Languages data docs50%
44Llama 3 8B Instruct· SeaLLMs Leaderboard for SEA Languages data docs49.4%
45Meta Llama 3 8b· SeaLLMs Leaderboard for SEA Languages data docs49%
46Mixtral 8X7b V0.1· SeaLLMs Leaderboard for SEA Languages data docs47.9%
47Mixtral 8x7B Instruct v0.1· SeaLLMs Leaderboard for SEA Languages data docs47.6%
48Qwen2.5 1.5B· SeaLLMs Leaderboard for SEA Languages data docs47.3%
49Qwen2.5 1.5B Instruct· SeaLLMs Leaderboard for SEA Languages data docs47.2%
50Gemma 2 2B IT· SeaLLMs Leaderboard for SEA Languages data docs45.9%

Showing 1–50 of 57 results.

Data credits: SeaExam and SeaBench leaderboard, SeaLLMs (DAMO-NLP-SG). Published leaderboard scores; results licence not stated.