# Ratecard

> Published AI API list prices (USD per 1M tokens), one row per model and provider, each with a source URL and check date. 85 rows, 78 verified, 14 providers. Last checked 2026-10-26.

## Methodology
Prices come only from each provider's own pricing page, docs, or public pricing API. Unverified means we could not read a number on the check date — the cell stays blank. We never invent rates. Details: https://ratecardai.pages.dev/methodology.html

## Data (CC BY 4.0 — please attribute "Ratecard" + link https://ratecardai.pages.dev/)
- [All models (JSON)](https://ratecardai.pages.dev/api/models.json)
- [All models (CSV)](https://ratecardai.pages.dev/api/models.csv)
- [Providers (JSON)](https://ratecardai.pages.dev/api/providers.json)
- [Change log (JSON)](https://ratecardai.pages.dev/api/changes.json)
- [Change log (CSV)](https://ratecardai.pages.dev/api/changes.csv)
- [OpenAPI](https://ratecardai.pages.dev/openapi.json)
- [Full dump](https://ratecardai.pages.dev/llms-full.txt)

## Dataset summary
- Units: USD per 1M tokens (unless unit/other says otherwise)
- Fields: input, output, cached, batch_input, batch_output, context, status, source, checked
- Cheapest verified text (in+out): OpenRouter Qwen3.7 Flash (Qwen) ($0.030 / $0.130)

## Cheapest verified text models (input+output per 1M)
- OpenRouter Qwen3.7 Flash (Qwen): in $0.030, out $0.130 (source https://openrouter.ai/qwen/qwen3.7-flash, 2026-10-26)
- Fireworks NVIDIA Nemotron 3.5 Lightning 30B A3B: in $0.050, out $0.200 (source https://docs.fireworks.ai/serverless/pricing, 2026-10-26)
- Together Llama 3 8B Instruct Lite: in $0.140, out $0.140 (source https://www.together.ai/pricing, 2026-10-26)
- Groq openai/gpt-oss-20b: in $0.075, out $0.300 (source https://console.groq.com/docs/models, 2026-10-26)
- OpenRouter Llama 4 Scout (Meta): in $0.100, out $0.300 (source https://openrouter.ai/meta-llama/llama-4-scout, 2026-10-26)
- OpenAI gpt-6-luna: in $0.100, out $0.500 (source https://platform.openai.com/docs/pricing, 2026-10-26)
- Anthropic Claude Haiku 5.5: in $0.100, out $0.500 (source https://docs.anthropic.com/en/docs/about-claude/pricing, 2026-10-26)
- Azure OpenAI GPT-6 Luna (short context, Global): in $0.100, out $0.500 (source https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/, 2026-10-26)
- Azure OpenAI GPT-6 Luna (short context, Data Zone): in $0.110, out $0.550 (source https://azure.microsoft.com/en-us/pricing/details/cognitive-services/openai-service/, 2026-10-26)
- Together gpt-oss-120B: in $0.150, out $0.600 (source https://www.together.ai/pricing, 2026-10-26)

## Pages
- [Calculator](https://ratecardai.pages.dev/calculator.html)
- [Compare](https://ratecardai.pages.dev/compare.html)
- [Developers / cite](https://ratecardai.pages.dev/developers.html)
- [Methodology](https://ratecardai.pages.dev/methodology.html)