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.
One or more sources degraded; showing last-known data. AA: degraded, retry after 41264s (AA API rate limited)
Fetched 2 Sept 2026, 18:14
Fetched 3 Sept 2026, 06:35
Published — · fetched —
Model table
| Links | ||||||
|---|---|---|---|---|---|---|
| Claude Fable 5.1 | Anthropic | 65.7 | $3.69 | $10.00 | $50.00 | OpenRouterLab |
| Claude Opus 5 | Anthropic | 63.1 | $2.34 | $5.00 | $25.00 | OpenRouterLab |
| Grok 4.6 | xAI | 60.9 | $0.9372 | $2.00 | $6.00 | OpenRouterLab |
| Kimi K3 | Moonshot AI | 59.7 | $0.8375 | $3.00 | $15.00 | OpenRouterLab |
| GLM 5.3 | Z.ai | 59.5 | $0.6829 | $1.40 | $4.40 | OpenRouterLab |
| Gemini 3.8 Flash | 58.7 | $0.5766 | $0.7500 | $3.75 | OpenRouterLab | |
| GLM 5.3 Flash | Z.ai | 57.5 | $0.0869 | $0.0750 | $0.2500 | OpenRouterLab |
| GPT-5.6 Luna | OpenAI | 52.3 | $0.0487 | $0.2000 | $1.20 | OpenRouterLab |
| Hunyuan Hy3 | Tencent | 42.2 | $0.0357 | $0.1320 | $0.5280 | OpenRouterLab |
| Llama 4 Maverick | Meta | 14.5 | $0.0346 | $0.2000 | $0.6960 | OpenRouterLab |
| Llama 4 Scout | Meta | 10.3 | $0.0106 | $0.1000 | $0.3000 | OpenRouterLab |
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".