Rates
Log scale — the only way to see $0.042 and $10 in one view. Jev's output bar is a hairline: outputs are free.
Bill explorer
Drag the sliders — every bar re-sorts instantly, cheapest first.
Scale
Monthly cost as input tokens scale from 100K to 100M (output fixed at half of input).
Source table
Every figure verified against the provider's official pricing page.
| Model | Input $/MTok | Output $/MTok | Context | Source |
|---|---|---|---|---|
| GPT-6.1 Sol OpenAI | $2.000 | $10.00 | ≈1M Standard short-context rate; requests over 272K prompt tokens bill at 2× input / 1.5× output. | Official pricing ↗ |
| Claude Sonnet 5.5 Anthropic | $2.000 | $10.00 | 1M | Official pricing ↗ |
| Grok 4.7 xAI | $2.000 | $6.00 | 500K Standard rate below 200K prompt tokens; long-context requests bill at 2×. | Official pricing ↗ |
| Gemini 3.8 Flash | $0.750 | $3.75 | 1M Introductory rate through Dec 31, 2026; $1.50 / $7.50 from Jan 1, 2027. | Official pricing ↗ |
| DeepSeek V4 Pro DeepSeek | $1.320 | $3.96 | 1M Peak hours; off-peak (nights & weekends UTC) is half. | Official pricing ↗ |
| Mistral Large 3 Mistral | $0.500 | $1.50 | 256K | Official pricing ↗ |
| Jev TypeSafe | $0.042 | FREE | — Early-access pricing from the official launch post; output tokens are free ("too cheap to meter"). | Official pricing ↗ |
Prices verified against official provider pricing pages on October 03, 2026. Planning estimates only — they ignore prompt-caching discounts, batch rates, tool-call charges, and regional premiums. Jev's outputs are free because they are tiny typed decisions, not generated text; it can't do what chat models do, so “cheaper” only applies to decision-shaped work.