Solutions · Analytics
LLM Pareto Frontier
Deterministic quality-vs-cost frontier for major LLMs, built from official structured data: Artificial Analysis v2, OpenRouter, and LM Arena's published Hugging Face dataset. No scraping, no fuzzy matching.
Fetched 30 Sept 2026, 21:26 UTC
Fetched 16 Oct 2026, 18:21 UTC
Published — · fetched —
Model table
| Links | ||||||
|---|---|---|---|---|---|---|
| Claude Opus 5.5 | Anthropic | 57.6 | $5.98 | $4.00 | $20.00 | OpenRouterAALab |
| GPT-6 Astra | OpenAI | 52.7 | $3.26 | $10.00 | $50.00 | OpenRouterAALab |
| GPT-6.1 Sol | OpenAI | 51 | $0.3929 | $2.00 | $10.00 | OpenRouterAALab |
| GLM 5.3 Flash | Z.ai | 41.8 | $0.2533 | $0.0136 | $0.7189 | OpenRouterAALab |
| GPT-5.6 Luna | OpenAI | 37.3 | $0.1783 | $0.2000 | $1.20 | OpenRouterAALab |
| GPT-6 Luna | OpenAI | 33.9 | $0.0421 | $0.1000 | $0.5000 | OpenRouterAALab |
| Gemini 3.1 Flash Lite Preview | 15.6 | $0.0394 | $0.2500 | $1.50 | OpenRouterAALab | |
| Command A+ | Cohere | 13.1 | $0.0000 | $0.3000 | $1.50 | OpenRouterAALab |
Methodology
Pareto dominance: model A dominates B iff Q(A) ≥ Q(B) and C(A) ≤ C(B) and at least one inequality is strict. Models missing the selected metric are excluded from that frontier. The graph and table show only the currently selected Pareto-efficient models.
Blended OpenRouter cost = inputShare × input $/1M + outputShare × output $/1M (visible and editable above when selected).
Arena WebDev uses Bradley-Terry Arena Scores (rating ± bounds, vote counts). Arena Agent uses IPS scores (score ± CI, observation and session counts). These methodologies are kept semantically separate and are never merged into a single "Elo".