Agentic benchmark
Toolathlon leaderboard
Every model the catalog carries a published Toolathlon value for, ranked by that value.
CategoryAgentic
MeasureInteractive tool-calling evaluation
TasksMulti-tool workflows
DifficultyAdvanced tool use
Published byIntroducing GPT-5.4 mini and nano
A tool-use benchmark focused on selecting, sequencing, and completing tasks with external tools.
Toolathlon ranking
22 models with a published Toolathlon value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Interactive tool-calling evaluation |
|---|---|---|---|
| 1 |  Muse Spark 1.1 | Meta | 75.6 |
| 2 |  Claude Opus 4.8 | Anthropic | 59.9 |
| 3 |  GPT-5.6 Sol | OpenAI | 58 |
| 4 |  Gemini 3.5 Flash | 56.5 | |
| 5 | GPT-5.5 | OpenAI | 55.6 |
| 6 | GPT-5.4 | OpenAI | 54.6 |
| 7 | GPT-5.6 Luna | OpenAI | 53.4 |
| 8 | GPT-5.6 Terra | OpenAI | 53.1 |
| 9 |  DeepSeek V4 Pro 0813 | DeepSeek | 51.8 |
| 10 |  Kimi K2.6 | Moonshot AI | 50 |
| 11 |  Step 3.7 Flash | StepFun | 49.5 |
| 12 |  GLM-5.2 | Z.AI | 48.2 |
| 13 | DeepSeek V4 Flash 0731 | DeepSeek | 47.8 |
| 14 |  MiniMax M2.7 | MiniMax | 46.3 |
| 15 | Claude Opus 4.5 | Anthropic | 43.5 |
| 16 | GPT-5.4 mini | OpenAI | 42.9 |
| 17 |  Qwen3.6 Plus | Alibaba | 39.8 |
| 18 | GLM-5 | Z.AI | 38 |
| 19 | Qwen3.5 397B | Alibaba | 36.3 |
| 20 | GPT-5.4 nano | OpenAI | 35.5 |
| 21 | Kimi K2.5 | Moonshot AI | 27.8 |
| 22 | Qwen3.6-35B-A3B | Alibaba | 26.9 |
Evidence key: Observed
Rows are ordered by the value Introducing GPT-5.4 mini and nano published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.