AI Benchmark Leaderboard
Compare AI model scores, coding ability and performance in one place.
| # | Model | Org | Intelligence | Coding | Math | MMLU Pro | GPQA | LiveCodeBench | AIME 2025 | MATH 500 | SciCode | IFBench | HLE |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 51 | JT-35B-Flash | China Mobile | 36.1 | 28.9 | — | — | 82.9 | — | — | — | 29.1 | 42.0 | 6.1 |
| 52 | Muse Ember (Max) | Meta | 36.0 | — | — | — | — | — | — | — | — | — | — |
| 53 | Agnes 3.0 Flash | Sapiens AI | 35.5 | — | — | — | 92.4 | — | — | — | 51.6 | — | 38.5 |
| 54 | Grok 4.7 (Xhigh) | SpaceXAI | 35.5 | — | — | — | — | — | — | — | — | — | — |
| 55 | Grok 4.7 (High) | SpaceXAI | 35.2 | — | — | — | — | — | — | — | — | — | — |
| 56 | Claude Opus 5.5 (Low, Default Fallback) | Anthropic | 34.8 | — | — | — | — | — | — | — | — | — | — |
| 57 | GPT-6.1 Sol (Low) | OpenAI | 34.8 | — | — | — | — | — | — | — | — | — | — |
| 58 | Claude Haiku 5.5 (Xhigh) | Anthropic | 34.6 | — | — | — | — | — | — | — | — | — | — |
| 59 | Tencent Hy4 | Tencent | 34.6 | — | — | — | — | — | — | — | — | — | — |
| 60 | Muse Ember (Xhigh) | Meta | 34.1 | — | — | — | — | — | — | — | — | — | — |
| 61 | JT-4.1 Flash 236B A21B (Reasoning) | China Mobile | 33.9 | — | — | — | — | — | — | — | 52.0 | — | 43.6 |
| 62 | Qwen3.8 Max (0902) | Alibaba | 33.7 | — | — | — | — | — | — | — | — | — | — |
| 63 | DeepSeek V4.1 Flash 1017 (Max) | DeepSeek | 32.6 | — | — | — | — | — | — | — | — | — | — |
| 64 | GPT-5.6 Terra (Max) | OpenAI | 32.3 | — | — | — | — | — | — | — | — | — | — |
| 65 | GLM-5.3 (Max) | Z AI | 31.7 | — | — | — | — | — | — | — | — | — | — |
| 66 | Grok 4.7 (Low) | SpaceXAI | 31.1 | — | — | — | — | — | — | — | — | — | — |
| 67 | Grok 4.6 (Xhigh) | SpaceXAI | 31.0 | — | — | — | — | — | — | — | — | — | — |
| 68 | Claude Sonnet 5.5 (Medium, Default Fallback) | Anthropic | 30.8 | — | — | — | — | — | — | — | — | — | — |
| 69 | Gemini 3.8 Flash (High) | 30.8 | — | — | — | — | — | — | — | — | — | — | |
| 70 | Inkling 2 Small | Thinking Machines | 30.6 | — | — | — | — | — | — | — | — | — | — |
| 71 | GPT-6.1 Terra (Non-reasoning) | OpenAI | 30.5 | — | — | — | — | — | — | — | — | — | — |
| 72 | Gemini 3.8 Flash (Medium) | 30.1 | — | — | — | — | — | — | — | — | — | — | |
| 73 | Claude Haiku 5.5 (High) | Anthropic | 29.9 | — | — | — | — | — | — | — | — | — | — |
| 74 | MiMo-V2.6-Pro | Xiaomi | 29.9 | — | — | — | — | — | — | — | — | — | — |
| 75 | Ling 3.1 Flash 1019 | InclusionAI | 29.6 | — | — | — | — | — | — | — | — | — | — |
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.