Quote a prompt
Token estimate is characters divided by 4. Each row is that estimate times the published list price. This page does not call a model.
Estimated input: 0 tokens. Tariff date: 2026-10-19.
| Model | Provider | List price | What the row is for |
|---|---|---|---|
| GPT-6 Luna | OpenAI | $0.000150 | High-volume cheap OpenAI — Cached input $0.01 / 1M |
| Claude Haiku 5.5 | Anthropic | $0.000150 | Fast Anthropic workhorse — Short-context rate for prompts up to 100k tokens |
| DeepSeek Flash | DeepSeek | $0.000360 | Lowest DeepSeek cost — Peak cache-miss; off-peak is half |
| Grok 4.3 | xAI | $0.000750 | Cheaper Grok tier |
| Gemini 3.8 Flash | $0.001125 | Cheapest capable Flash — Promo through December 31, 2026; then $1.50 / $7.50 | |
| DeepSeek V4 Pro | DeepSeek | $0.001188 | Strong value reasoning — Peak cache-miss; off-peak is half |
| Grok 4.7 | xAI | $0.001800 | xAI flagship general |
| GPT-6.1 Sol | OpenAI | $0.003000 | Daily driver coding & chat — Cached input $0.10 / 1M |
| Claude Sonnet 5.5 | Anthropic | $0.003000 | Coding & agents balance |
| Gemini 3.1 Pro | $0.003600 | Multimodal / long context — <=200k prompt tokens | |
| Claude Opus 5.5 | Anthropic | $0.006000 | Hard agentic coding |
| GPT-6 Astra (baseline) | OpenAI | $0.015000 | Flagship reasoning / agents — Cached input $1 / 1M |
| Claude Fable 5.1 | Anthropic | $0.015000 | Long-running agents |
gpt-6-astra for this estimate is $0.015000. The gap to a cheaper row is the difference in list price, not a measured saving.