Gestalt (stealth) Model Pricing & API Watch: 1,928 Elo vs Claude Fable 5.2 (1,714) on Frontier Design Benchmark (2026)
Day-0 tracker for Gestalt (stealth), the unattributed stealth AI model dominating Contra Labs Frontier Design Benchmark (FDB) at 1,928 Elo (~89%โ90% win rate, preferred 78% of the time over Claude Fable 5.2 at 1,714 Elo): current attribution hypotheses (Claude Fable 5.5, Gemini 4 Xenon/Krypton, GLM 5.5), expected frontend/design token cost tiers, and live API availability watch.
Live 2026 FinOps Benchmark Matrix: Day-0 Stealth Model & Pricing Watch
| Architecture / Model Tier | 2026 Unit Economics & Pricing | Production FinOps Impact |
|---|---|---|
| FDB Leaderboard Score & Win Rate | 1,928 Elo (~89%โ90% win rate) ยท +214 Elo ahead of Claude Fable 5.2 (1,714) | Outperformed human golden reference designs in blind Frontier Design Benchmark evaluations |
| API Availability & Pricing Status | Stealth / Unattributed Preview Only โ $0 Public API Yet | TokenMargin 30-minute Day-0 Radar monitors OpenRouter, Anthropic, Google & Zhipu feeds |
| Attribution Hypotheses | Speculated as Claude Fable 5.5, Gemini 4 Xenon/Krypton, or GLM 5.5 | Expected production tier: $2โ$15/1M input if frontier proprietary, or <$1/1M if open-weight MoE |
| Frontend & Agent Design Token Math | UI/HTML/CSS + React component generation averages 4,000โ12,000 output tokens/turn | High output-to-input ratio makes output $/1M rate and prefix caching critical at 10K DAU |
Related FinOps Playbooks & Model Cost Calculators
Drop Your Usage CSV, Paste Your System Prompt, or Rescue Your Monthly LLM Bill
100% browser-local FinOps auditing: drop your OpenRouter / OpenAI / Anthropic / Cursor usage CSV, count 20-turn agent loop bloat from your system prompt, or model instant savings across 61 models.
Export your 30-day activity CSV from OpenRouter, Anthropic Console, OpenAI Usage, or Cursor/Cline and drop it below (or paste raw CSV rows). Everything runs in your browser memory โ no API keys or logs ever leave your device.