AI evaluation startup LMArena has raised $150 million in Series A funding, SiliconANGLE reported on Jan. 6 (local time).
The round included Felicis, UC Investments, Andreessen Horowitz and Kleiner Perkins. LMArena was valued at $1.7 billion in the round. That is triple the figure from 7 months earlier.
LMArena operates a cloud platform that evaluates AI model performance. Typical AI benchmarks use static questions, but LMArena uses real-time crowdsourced prompts for a more accurate assessment. Users compare answers generated by 2 AI models and provide feedback, which can be used to rank AI model performance.
Google Gemini 3 Pro is currently rated as the top-performing model. OpenAI also tested GPT-5 on the LMArena platform before its release.
About the Author
Chi-kyu Hwang
delight@d-today.co.kr