LLM Trends & Benchmarks
Track the evolution of AI models over time. Compare benchmark scores, pricing trends, and the open source vs closed source frontier.
Benchmark vs Price
How does model performance correlate with API pricing? Higher benchmark scores with lower prices indicate better value.
MMLU-Pro vs Price
GPQA vs Price
Aider (Coding) vs Price
LiveCodeBench vs Price
MATH Hard vs Price
Context Length vs Price
Benchmarks Over Time
Track the frontier of AI capabilities as new models are released each month.
MMLU-Pro Frontier Over Time
Provider frontier comparison - MMLU-Pro benchmark (includes reasoning variants)
GPQA Frontier Over Time
Provider frontier comparison - Graduate-level science (includes reasoning variants)
Aider Frontier Over Time
Provider frontier comparison - Real-world coding (includes reasoning variants)
LiveCodeBench Frontier Over Time
Provider frontier comparison - Competitive programming (includes reasoning variants)
MATH Hard Frontier Over Time
Provider frontier comparison - Competition math (includes reasoning variants)
Context Length Frontier Over Time
Provider frontier comparison - Maximum context window
Open Source vs Closed Source
Compare the frontier capabilities between open source and proprietary models over time.
MMLU-Pro: Open vs Closed
Comparing the best open source and closed source models each month
GPQA: Open vs Closed
Graduate-level science benchmark comparison
Aider: Open vs Closed
Real-world coding benchmark comparison
LiveCodeBench: Open vs Closed
Competitive programming benchmark comparison
MATH Hard: Open vs Closed
Competition math benchmark comparison
Context Length: Open vs Closed
Maximum context window comparison
Data includes 296 models from 2023-05 to 2025-12
Benchmark data sourced from Artificial Analysis. Pricing data updated daily.