EyesInAI — AI Benchmark Leaderboard

⚡Ping

Latency & availability — single-word reply

364 ms

🧮Reasoning

Basic math reasoning — show work, give answer

443 ms

{ }JSON Output

Structured output compliance — valid JSON with required keys

435 ms

💻Code Gen

Python function generation with docstring

677 ms

🚀Throughput

Token generation speed — 500-token long-form response

1766 ms

🔍Context Recall

Retrieval from in-context data — 20-item list Q&A

457 ms

🔧Tool Use

Function/tool calling — get_weather invocation

310 ms

📝Summarization

Real-world: distill a news article into 3 bullet points

558 ms

🏷️Classification

Real-world: sentiment + category from customer review

275 ms

📄Data Extraction

Real-world: pull structured fields from an invoice

441 ms

📋Instruction Follow

Real-world: multi-rule compliance (5 sentences, no "the", etc.)

579 ms

llama-3.3-70b-versatile