AI Benchmark Leaderboard

Compare AI model scores, coding ability and performance in one place.

# Model Org Intelligence ↓ CodingMathMMLU ProGPQALiveCodeBenchAIME 2025MATH 500SciCodeIFBenchHLE
101 Gemini 3.8 Flash (Low) Google 22.5 ——————————
102 GPT-6 Luna (High) OpenAI 21.4 ——————————
103 K2 Horizon 7B MBZUAI Institute of Foundation Models 21.0 38.6 —— 75.3 ——— 27.5 — 18.2
104 DeepSeek V4 Pro 0813 (High) DeepSeek 20.8 ——————————
105 Nanbeige4.3-33B Nanbeige 20.4 ——————————
106 Qwen3.8 27B (Xhigh) Alibaba 20.1 ——————————
107 JT-4.1 Flash 236B A21B China Mobile 19.0 ——————————
108 DeepSeek V4 Flash (Non-reasoning) DeepSeek 18.9 ——— 71.6 ———— 47.2 7.8
109 MiniMax-M3 MiniMax 18.9 ——————————
110 Claude Haiku 5.5 (Low) Anthropic 18.6 ——————————
111 DeepSeek V4.1 Flash 1017 (Non-reasoning) DeepSeek 18.5 ——————————
112 GPT-6 Luna (Medium) OpenAI 18.5 ——————————
113 K2 Horizon 375B A23B Institute of Foundation Models 17.8 ——————————
114 Beam 1.0 Reflection 17.7 ——————————
115 Qwen3.8 27B (Medium) Alibaba 17.5 ——————————
116 Kimi K3 (Low) Kimi 17.4 ——————————
117 Ministral 14B (Dec '25) Mistral 16.3 10.7 30.0 69.3 57.2 35.1 30.0 — 23.6 32.0 4.6
118 K2 Horizon 3.7B MBZUAI Institute of Foundation Models 16.2 26.1 —— 69.2 ——— 22.0 — 13.9
119 Qwen3.8 27B (Low) Alibaba 15.7 ——————————
120 Step3 VL 10B StepFun 15.4 13.9 —— 69.0 ——— 31.1 50.2 10.2
121 Ministral 8B (Dec '25) Mistral 15.1 9.8 31.7 64.2 47.1 30.3 31.7 — 20.8 29.1 4.3
122 Phi-3 Medium Instruct 14B Microsoft Azure 14.4 8.9 1.3 54.3 32.6 15.0 1.3 46.3 11.8 21.5 4.5
123 Solar Pro 4 Upstage 14.2 ——————————
124 ZAYA1-8B Zyphra 14.1 5.6 —— 47.1 ——— 12.3 48.6 4.2
125 Apodex 1.1 Apodex 13.7 ——————————

AI Benchmarks FAQ

Which AI model is best for my task?

Compare the scores most relevant to your task, then consider speed, pricing and practical requirements. A high score in one benchmark does not make a model the best choice for every use.

What does a missing benchmark score mean?

A dash means that a score is not available for that model in this table. It should not be treated as a score of zero.

How are the top models ranked?

By the intelligence index, highest first. The Top models cards show the eight highest-scoring current models, one per model family, and models with the same score share a place.

Where do the scores come from?

From independent third-party model evaluations, which Franklin AI syncs automatically each day. Each model page links to About these benchmarks, which explains them in more detail.

What does a model page show?

Its scores, its price per million input and output tokens, its measured speed and comparisons with other models, where that data is available.