OJBench

OJBench - competitive programming and reasoning benchmark.

Category

STEM & Reasoning

Max Score

—

Score Type

percent

Active

Yes

Model Scores

Scores for OJBench

Model Score Date Verified
Qwen3.5-4B
31.57%
01.02.2026 Verified
Qwen3.5-122B-A10B
54.19%
01.02.2026 Verified
Qwen3.5-27B
55.07%
01.02.2026 Unverified
GPT-5-mini 2025-08-07
55.51%
01.02.2026 Unverified
Claude Opus 4.6
84.73%
23.08.2026 Verified
Gemini 3.1 Pro
100.00%
23.08.2026 Verified
Kimi K2.5
76.51%
23.08.2026 Verified
Gemini-2.5-Flash-Thinking
30.69%
31.07.2025 Verified
Qwen3-30B-A3B
26.58%
29.04.2025 Verified
Qwen3-30B-A3B-Thinking-2507
33.04%
31.07.2025 Verified
LFM2.5-2.6B
12.63%
— Verified
Qwen3.5-2B
2.60
— Verified
granite-4.2-3B
28.19%
— Verified
Nemotron-3-Nano-4B
25.55%
— Verified
Gemma4-E4B
24.08%
— Verified
LFM2.5-8B-A1B
8.22%
— Verified
Qwen3.5-35B-A3B
49.05%
01.02.2026 Unverified
Qwen3-235B-A22B
44.20%
01.02.2026 Unverified
Kimi K2.6
85.17%
23.08.2026 Verified
MiniCPM5-2B
43.91%
— Verified
Gemma4-E2B
13.22%
— Verified
Qwen3-Next-80B-A3B-Thinking
39.79%
10.09.2025 Verified