Inference

Toolathlon-Verified

A verified tool-use benchmark variant for completing multi-step workflows with external tools.

  • In index
  • Agentic
ModelsModels with a result.20
Top resultBest result on this test.80.6%Claude Opus 5
Top-3 spreadPoints from first to third.2.8 pts
YearYear of release.N/A

Result and price

020406080100$0.1$0.3$1$3$10$30$100RESULTOUTPUT PRICE PER 1M TOKENS · LOG SCALE

Results

20 results
#ModelResultThe score from the source.
01Claude Opus 580.6%
02GLM-5.3-Flash78.4%
03Claude Opus 5.577.8%
03Claude Fable 5.177.8%
05MiMo-V2.6-Pro76.9%
06Step 5 Preview74.1%
06Hy4 preview74.1%
06DeepSeek V4 Pro 081374.1%
09MiMo-V2.6-Flash73.6%
10Qwen3.8-Flash-Next73.5%
11Kimi K373.2%
12GLM-5.373.0%
13Qwen3.8 Max72.5%
14Ornith-1.5-397B71.2%
15DeepSeek V4 Flash 073170.3%
16dots3-note Preview55.6%
17Inkling-Small54.4%
18Laguna S 2.149.7%
19Ornith-1.5-35B-A3B48.7%
20Ornith-1.5-9B41.2%