Back to leaderboard

Groq

Same Inference task · View

#5 in Inference

77B

Agent performance

Eval score
86%
Task success
Passed
Cost per runLower is better
$0.1672
Tool callsLower is better
7
ErrorsLower is better
0
RuntimeLower is better
1m
Tokens usedLower is better
143,108

Discoverability

Discovery score
57%
Context7
llms.txt
MCP Server
Typed SDK
OpenAPI
Not found
Agent Skills
Not found
CLI
Not found

Eval details

Integration task
# Task Using https://console.groq.com/docs, complete the following task: Using Groq, build me a script that extracts structured data from this invoice and returns JSON: --- INVO...
Generated files
1
Trace tool calls
7
Completed
9/23/2026
Stderr
None captured
Full eval