Home
Blog
tool use

tool use

5 articles tagged “tool use”.

Concepts
Agentic Evals: How We Benchmark Tool-Using LLMs
8 min read
Benchmark explained
What Is MCP-Atlas? Scaled Tool Use Explained
7 min read
Comparison
The Best LLM for Tool Use and Function Calling in 2026
7 min read
Concepts
What Is an LLM Agent? Tools, Planning and Evaluation
8 min read
Comparison
The Best LLM for AI Agents in 2026
8 min read

Benchmarks Compare Models Blog Glossary

About these benchmarks

This page compares leading frontier AI models across the evaluations labs use to measure real-world ability — agentic coding, terminal and computer use, tool orchestration, web search, long-context and graduate-level reasoning, visual understanding and multilingual knowledge. Choose any two models above to see a head-to-head: the first becomes the baseline and the second is scored green where it wins and red where it falls behind, benchmark by benchmark.

Figures are drawn from each model’s published system card and from independent leaderboards such as Artificial Analysis. Where a lab does not report a given benchmark, the cell is left blank (—) rather than estimated.

Explore each benchmark

Agentic coding · SWE-bench Pro
Agentic coding · SWE-bench Verified
Long context reasoning · AA-LCR
Agentic terminal coding · Terminal-Bench 2.1
Multidisciplinary reasoning · Humanity's Last Exam
Agentic search · BrowseComp
Scaled tool use · MCP-Atlas
Agentic computer use · OSWorld-Verified
Agentic financial analysis · Finance Agent v2
Cybersecurity vulnerability reproduction · CyberGym
Graduate-level reasoning · GPQA Diamond
Visual reasoning · CharXiv Reasoning
Multilingual Q&A · MMMLU

From the blog

All articles →

The Complete Guide to LLM Benchmarks
How to Evaluate an LLM for Your Own Use Case
How to Choose an LLM: A Benchmark-Driven Framework
Why LLM Benchmarks Saturate (and What Comes Next)
What Is MMLU-Pro? A Harder Knowledge Benchmark
What Is Humanity’s Last Exam? The Frontier Reasoning Benchmark

Benchmark data is provided for informational comparison and may be updated as labs publish new results.

About Methodology