Hong Life CalculatorsHow it works
For every model in our price file the monthly cost is worked out with the same formula as the single-provider calculators: (uncached input tokens × input price + cached tokens × cached-input price + output tokens × output price) ÷ 1,000,000 × requests per day × days per month, with the provider’s official batch or off-peak discount applied to the share of requests you choose. The list is sorted from cheapest to most expensive; use “Show ranks” to page through it. Prices were checked on 10 Nov 2026 on each provider’s official pricing page and change often. The ranking is price only — it says nothing about quality, speed or which model fits your task.
All prices used (US$ per 1M tokens, checked 10 Nov 2026)
| Provider | AI model | Input price | Cached-input price | Output price | Official discount |
|---|---|---|---|---|---|
| OpenAI | gpt-6-astra | 10 | 1 | 50 | 50% |
| OpenAI | gpt-6.1-sol | 2 | 0.1 | 10 | 50% |
| OpenAI | gpt-6.1-terra | 1 | 0.05 | 5 | 50% |
| OpenAI | gpt-6-sol | 2 | 0.2 | 10 | 50% |
| OpenAI | gpt-6-luna | 0.1 | 0.01 | 0.5 | 50% |
| OpenAI | gpt-5.6-sol | 4 | 0.4 | 20 | 50% |
| OpenAI | gpt-5.6-terra | 2 | 0.2 | 12 | 50% |
| OpenAI | gpt-5.6-luna | 0.2 | 0.02 | 1.2 | 50% |
| OpenAI | gpt-5.5 | 5 | 0.5 | 30 | 50% |
| OpenAI | gpt-5.4 | 2.5 | 0.25 | 15 | 50% |
| OpenAI | gpt-5.4-mini | 0.75 | 0.075 | 4.5 | 50% |
| OpenAI | gpt-5.4-nano | 0.2 | 0.02 | 1.25 | 50% |
| OpenAI | gpt-5-mini | 0.25 | 0.025 | 2 | 50% |
| OpenAI | gpt-5-nano | 0.05 | 0.005 | 0.4 | 50% |
| OpenAI | gpt-4.1 | 2 | 0.5 | 8 | 50% |
| OpenAI | gpt-4.1-mini | 0.4 | 0.1 | 1.6 | 50% |
| OpenAI | gpt-4o-mini | 0.15 | 0.075 | 0.6 | 50% |
| Anthropic | Claude Fable 5.2 | 8 | 0.2 | 40 | 50% |
| Anthropic | Claude Mythos 5.2 | 8 | 0.2 | 40 | 50% |
| Anthropic | Claude Opus 5.5 | 4 | 0.2 | 20 | 50% |
| Anthropic | Claude Sonnet 5.5 | 2 | 0.1 | 10 | 50% |
| Anthropic | Claude Haiku 5.5 (≤100K prompt) | 0.1 | 0.01 | 0.5 | 50% |
| Anthropic | Claude Opus 5 | 5 | 0.5 | 25 | 50% |
| Anthropic | Claude Sonnet 5 | 2 | 0.2 | 10 | 50% |
| Anthropic | Claude Opus 4.8 | 5 | 0.5 | 25 | 50% |
| Anthropic | Claude Sonnet 4.6 | 3 | 0.3 | 15 | 50% |
| Anthropic | Claude Haiku 4.5 | 1 | 0.1 | 5 | 50% |
| Gemini 3.6 Flash | 1.5 | 0.15 | 7.5 | 50% | |
| Gemini 3.5 Flash | 1.5 | 0.15 | 9 | 50% | |
| Gemini 3.5 Flash-Lite | 0.3 | 0.03 | 2.5 | 50% | |
| Gemini 3.1 Pro Preview (≤200K) | 2 | 0.2 | 12 | 50% | |
| Gemini 3.1 Flash-Lite | 0.25 | 0.025 | 1.5 | 50% | |
| Gemini 3 Flash Preview | 0.5 | 0.05 | 3 | 50% | |
| Gemini 2.5 Pro (≤200K) | 1.25 | 0.125 | 10 | 50% | |
| Gemini 2.5 Flash | 0.3 | 0.03 | 2.5 | 50% | |
| Gemini 2.5 Flash-Lite | 0.1 | 0.01 | 0.4 | 50% | |
| Meta | Muse Wonder 1.0 | 4 | 0.2 | 20 | — |
| Meta | Muse Spark 1.3/1.2/1.1 | 1.25 | 0.15 | 4.25 | — |
| Mistral AI | Mistral Large 4 | 0.7 | 0.07 | 2.1 | — |
| Mistral AI | Mistral Small 4 | 0.15 | 0.015 | 0.6 | — |
| Mistral AI | Ministral 3 14B | 0.2 | 0.02 | 0.2 | — |
| Mistral AI | Ministral 3 8B | 0.15 | 0.015 | 0.15 | — |
| Mistral AI | Ministral 3 3B | 0.1 | 0.01 | 0.1 | — |
| DeepSeek | deepseek-flash (V4.1) | 0.3 | 0.006 | 1.2 | 50% |
| DeepSeek | deepseek-v4-pro | 1.32 | 0.044 | 3.96 | 50% |
| xAI | grok-4.7 | 2 | 0.5 | 6 | 20% |
| xAI | grok-4.6 | 2 | 0.5 | 6 | 20% |
| xAI | grok-4.5 | 2 | 0.3 | 6 | 20% |
| xAI | grok-4.3 | 1.25 | 0.2 | 2.5 | 20% |
| xAI | grok-4.20 reasoning | 1.25 | 0.2 | 2.5 | 20% |
| xAI | grok-4.20 non-reasoning | 1.25 | 0.2 | 2.5 | 20% |
| xAI | grok-build-0.1 | 1 | 0.2 | 2 | — |
⚠ Estimates for information only — not financial advice.
FAQ
Why are some providers missing?
We only include a price we could read on the provider’s own pricing page. Cohere’s page showed only legacy prices and Amazon’s Nova prices did not load, so they are left out until they can be confirmed.
Is the cheapest model the best choice?
Not necessarily. A cheaper model that needs retries, longer prompts or human fixes can cost more overall. Test the two or three cheapest models that meet your quality bar on real requests.
Checked 10 Nov 2026. Official pricing pages of OpenAI, Anthropic, Google (Gemini Developer API), Meta (Meta Model API), Mistral AI, DeepSeek and xAI, read on 10 Nov 2026. Cohere and Amazon Nova are not included because current per-token prices could not be confirmed on their official pages.