Access the world's leading insights on AI
Artificial Analysis is the leading independent AI benchmarking company. We help you make decisions around AI and stay up to date across models, inference, and hardware.
- Build with our API
- Export data for local analysis
- Build custom charts and tables
- Access industry reports and guides
- Get email support
- For individuals and small organizations
- Everything in Pro
- Benchmark models, inference, and hardware for your use cases
- Highest API rate limits
- Compute market model access with forecasts
- Workshops and education
- AI strategy advisory
- Personalized support
- For organizations with 150+ employees
For data redistribution or external use, contact us.
Publicly Referenced By
Compare all features
| Feature | Pro | Enterprise |
|---|---|---|
| Tools | ||
| Data Playground | ||
| Table Builder | ||
| Data Export | ||
| API and Databook Download (See API Docs) | ||
| Language Model Data | ||
| Media Model Data (Image, Video, Speech, Music) | ||
| Language Model Provider Data | ||
| Reports | ||
| State of AI Report | ||
| AI Adoption Survey | ||
| Model Deployment Report | ||
| Leaders AI Strategy Guide | ||
| Support Contact | ||
| Email Support | ||
| Dedicated Support | ||
| Custom Support | ||
| Workshops and Education | ||
| AI Advisory | ||
| Custom Model, Inference, and Hardware Benchmarking | ||
| Compute Market Model with Supply, Demand Forecasts | ||
0+
Models Benchmarked
0+
Inference Providers
0+
Model-Host Pairs
0B+
Evaluation Tokens
Where decision-makers go for AI performance data
Just off stage at #GoogleIO — Gemini 3.5 Flash is available today across our products and APIs. Compared to 3.1 Pro it’s better across almost all benchmarks with huge progress in coding, and ~4x faster tokens/second than other frontier models…
Artificial Analysis was recently featured in Altimeter Capital and Bill Gurley’s latest BG² podcast with Sam Altman and Satya Nadella. Our benchmarks show the cost of accessing GPT-4 level intelligence falling ~350X over the past ~2 years. Further, we show the cost of accessing new frontiers of…
Jensen Huang’s keynote at Computex used Artificial Analysis benchmarks to communicate the performance of NVIDIA’s new Nemotron 3 Ultra model — featuring our Intelligence Index vs. Output Speed chart and GDPval-AA, our benchmark built on OpenAI’s GDPval dataset…
피할 수 없으면 즐겨라 — Korea has now established itself as a top 3 nation in the AI field. With support from the Korean government’s AI initiatives, several Korean AI research institutes with cutting-edge artificial intelligence have emerged.
Shoutout to the team that built https://artificialanalysis.ai. Really neat site that benchmarks the speed of different LLM API providers to help developers pick which models to use. This nicely complements LMSYS Chatbot Arena, Hugging Face open LLM leaderboards and Stanford's HELM...
Fireworks AI hits top performance on the Artificial Analysis leaderboard with Kimi K2 on NVIDIA GB200 NVL72 systems. Fireworks AI CEO Lin Qiao explains why Blackwell NVL72 transforms massive MoE serving.
Grok 4.1 Fast with reasoning just beat Gemini 3!
