Glossary / Benchmark

Benchmark

A standard test used to compare AI models, such as a set of coding tasks or exam questions. A high score does not always mean better real-world performance.

Also written as: benchmarks

In the news

• Databricks introduces an autonomous Genie for ongoing enterprise work20m ago
• Google DeepMind introduces Gemini Robotics 2.5 for more robot tasks1d ago
• OpenAI launches GPT-6 Cyber with new safeguards6d ago
• Together AI introduces compiler that continually optimizes AI inferenceNov 4
• GLM 5.3 built Firefox exploit chains in patch-gap testsOct 30
• NVIDIA releases Nemotron 3.5 Transcribe and Granary v2 on Hugging FaceOct 28
• OpenAI seeks up to $30B at a $1.25T valuationOct 23
• Google introduces Agent Compute Engine and expands agent governanceOct 22
• Symantec finds AI attack agent targeting exposed language modelsOct 22
• Nvidia expands open AI models for business data, robotics and scienceOct 21
• Inception releases Mercury 3 with faster results in its own testsOct 21
• DeepSeek releases DeepInfer inference framework as open sourceOct 21