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+

27 models

Top 5 Models Performance

Prism ML: ternary-bonsai-2-27b ######################################## 95.12
Prism ML: ternary-bonsai-27b ####################################### 93.9
orcarouter: orcacyber-nano-1.0 ####################################### 93.29
Mistral AI: mistral-small-3.2-24b-instruct-2506 ####################################### 92.9
Tencent: youtu-llm-2b ##################################### 89
– 970 – 1.1T
– 2019 – 2026
Rank Model Score
❓ 1 Prism ML: ternary-bonsai-2-27b 95.12
❓ 2 Prism ML: ternary-bonsai-27b 93.9
❓ 3 orcarouter: orcacyber-nano-1.0 93.29
🇫🇷 4 Mistral AI: mistral-small-3.2-24b-instruct-2506 92.9
🇨🇳 5 Tencent: youtu-llm-2b腾讯控股有限公司 89
❓ 6 webai-official: twil-8b 88.3
🇨🇳 7 Qwen: Qwen3 4B通义千问 87.8
🇨🇳 8 Qwen: qwen2.5-coder-7b-instruct通义千问 84.1
❓ 9 essentialai: rnj-1-instruct 83.5
🇺🇸 10 OpenAI: gpt-oss-20b 81.7
🇨🇳 11 Qwen: Qwen3 8B通义千问 81.7
❓ 12 Prism ML: ternary-bonsai-8b 77.4
🇨🇳 13 Qwen: qwen3-1.7b通义千问 76.2
🇷🇺 14 Sber: gigachat3.5-432b-a28b-baseСбер 75.61
🇨🇳 15 Xiaomi: MiMo-V2.5 Pro小米集团 75.6
🇺🇸 16 Hugging Face Smol Models Research: smollm3-3b 74.7
❓ 17 Prism ML: bonsai-8b 73.8
🇦🇪 18 Technology Innovation Institute: falcon-h1-1.5b-deep-instruct 68.9
🇺🇸 19 Apple: diffucoder-7b-instruct 65.2
🇦🇪 20 Technology Innovation Institute: falcon-h1-1.5b-instruct 61.59
🇺🇸 21 Apple: diffucoder-7b-base 60.4
🇨🇳 22 DeepSeek: deepseek-r1-distill-qwen-1.5b深度求索 59.5
🇨🇳 23 Qwen: qwen3-8b-base通义千问 58.5
❓ 24 essentialai: rnj-1 57.9
🇦🇪 25 Technology Innovation Institute: falcon-h1-1.5b-deep-base 46.34
🇦🇪 26 Technology Innovation Institute: falcon-h1-1.5b-base 42.68
🇺🇸 27 Arcee AI: trinity-nano-base 31.7