LLM API pricing comparison — every major model, one table
Standard pay-as-you-go API prices for 17 tracked models, cheapest input first, verified 2026-09-02. Every model links to a detail page with worked examples; the token cost calculator prices your own prompts, and the timeline shows how these prices keep collapsing.
| model | provider | input $/Mtok | output $/Mtok | chat msg cost |
|---|---|---|---|---|
| GPT-5.6 Luna | OpenAI | $0.2 | $1.2 | $0.0008 |
| GPT-5.4 nano | OpenAI | $0.2 | $1.25 | $0.000825 |
| Gemini 2.5 Flash | $0.3 | $2.5 | $0.00155 | |
| DeepSeek V4 Flash | DeepSeek | $0.44 | $1.32 | $0.0011 |
| Gemini 3 Flash | $0.5 | $3 | $0.002 | |
| Gemini 3.7 Flash | $0.75 | $3.75 | $0.00262 | |
| DeepSeek V4 Pro | DeepSeek | $1.32 | $3.96 | $0.0033 |
| GPT-5.6 Terra | OpenAI | $2 | $12 | $0.008 |
| Claude Sonnet 5 | Anthropic | $2 | $10 | $0.007 |
| Grok 4.5 | xAI | $2 | $6 | $0.005 |
| Grok 4.6 | xAI | $2 | $6 | $0.005 |
| Kimi K3 | Moonshot | $3 | $15 | $0.0105 |
| GPT-5.6 Sol | OpenAI | $4 | $20 | $0.014 |
| GPT-5.5 | OpenAI | $5 | $30 | $0.02 |
| Claude Opus 4.8 | Anthropic | $5 | $25 | $0.0175 |
| Claude Fable 5 | Anthropic | $10 | $50 | $0.035 |
| Claude Fable 5.1 | Anthropic | $10 | $50 | $0.035 |
Chat msg = 1K input + 500 output tokens. Standard tier, cache-miss; batch and promo discounts excluded. The spread between the cheapest and priciest tracked model is 44× on the same request.
The weekly e/acc newsletter
One email a week: what accelerated.
Frontier releases, price drops, compute buildouts — the week's acceleration in five minutes, sourced and numeric. Free.