Note: Overall leaderboard rankings may not reflect true model quality — individual benchmarks give a clearer picture. ARC-Challenge MMLU GPQA GSM8K Artificial Analysis Intelligence Index v4.1
← Back to leaderboard

HumanEval

34 models

Top 5 Models Performance

Tencent: youtu-llm-2b ######################################## 95.9
Qwen: Qwen3 4B ######################################## 95.4
webai-official: twil-8b ####################################### 93.2
OpenAI: gpt-4o ###################################### 90.2
Inception: mercury ###################################### 90
– 970 – 1.1T
– 2019 – 2026
Rank Model Score
🇨🇳 1 Tencent: youtu-llm-2b腾讯控股有限公司 95.9
🇨🇳 2 Qwen: Qwen3 4B通义千问 95.4
❓ 3 webai-official: twil-8b 93.2
🇺🇸 4 OpenAI: gpt-4o 90.2
🇺🇸 5 Inception: mercury 90
🇨🇳 6 DeepSeek: deepseek-v2.5深度求索 89
🇨🇳 7 Qwen: qwen3-1.7b通义千问 84.8
🇷🇺 8 Sber: gigachat3.5-432b-a28b-baseСбер 80.49
🇺🇸 9 Hugging Face Smol Models Research: smollm3-3b 79.9
🇦🇪 10 Technology Innovation Institute: falcon-h1-1.5b-deep-instruct 73.78
🇺🇸 11 Apple: diffucoder-7b-instruct 72
🇷🇺 12 Yandex: YandexGPT Lite 5Яндекс 71.8
🇦🇪 13 Technology Innovation Institute: falcon-h1-1.5b-instruct 68.29
🇺🇸 14 OpenAI: GPT-3.5 Turbo 68
🇺🇸 15 Apple: diffucoder-7b-base 67.1
🇷🇺 16 Yandex: gpt-5-lite-pretrainЯндекс 66.5
🇨🇳 17 Qwen: qwen2.5-coder-32b通义千问 65.9
🇨🇳 18 DeepSeek: deepseek-r1-distill-qwen-1.5b深度求索 64
🇨🇳 19 Qwen: qwen2.5-coder-14b通义千问 64
🇨🇳 20 Qwen: qwen2.5-coder-7b通义千问 61.6
🇦🇪 21 Technology Innovation Institute: falcon-h1-1.5b-deep-base 52.44
🇨🇳 22 Qwen: qwen2.5-coder-3b通义千问 52.4
🇦🇪 23 Technology Innovation Institute: falcon-h1-1.5b-base 50
🇺🇸 24 Google: Gemma 3 27B Pretrained 48.8
🇨🇳 25 Moonshot AI: Moonlight Instruct月之暗面 48.1
🇺🇸 26 Google: gemma-3-12b-pt 45.7
🇨🇳 27 Qwen: qwen2.5-coder-1.5b通义千问 43.9
🇨🇳 28 OpenBMB: bitcpm4-1b 37.2
🇺🇸 29 Arcee AI: trinity-nano-base 36.3
🇺🇸 30 Google: gemma-3-4b-pt 36
🇨🇳 31 Qwen: qwen-14b通义千问 32.3
🇨🇳 32 Qwen: qwen-7b通义千问 29.9
🇨🇳 33 Qwen: qwen2.5-coder-0.5b通义千问 28
🇨🇳 34 Qwen: qwen-1.8b通义千问 17.1