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

OCRBench-v2

6 models

Top 5 Models Performance

NVIDIA Corporation: cosmos3-super ######################################## 66.7
Qwen: qwen3-vl-32b-instruct ######################################## 65.9
Google: Gemma 4 31B Instruct ##################################### 61.8
Prism ML: ternary-bonsai-27b ##################################### 61.42
NVIDIA Corporation: cosmos-reason2-32b ################################## 57.4
– 970 – 1.1T
– 2019 – 2026
Rank Model Score
🇺🇸 1 NVIDIA Corporation: cosmos3-super 66.7
🇨🇳 2 Qwen: qwen3-vl-32b-instruct通义千问 65.9
🇺🇸 3 Google: Gemma 4 31B Instruct 61.8
❓ 4 Prism ML: ternary-bonsai-27b 61.42
🇺🇸 5 NVIDIA Corporation: cosmos-reason2-32b 57.4
❓ 6 Prism ML: ternary-bonsai-2-27b 56.88