AI Models

Foundation models on Amazon Bedrock are billed per token, and the same prompt can cost ten times more or less depending on the model, the tier, and whether you cache or batch. Start with a tool, a model maker, or the kind of work you need done.

Tools

Look up a rate, estimate a bill, or settle a close call between models.

Browse & Sort

Model Catalog

Every Bedrock foundation model in one sortable table: on-demand, batch, and prompt-cache rates per 1M tokens, with filters for maker, tier, and modality.

Open the catalog →
Estimate

Token Cost Calculator

Turn monthly prompt and completion volume into a dollar figure for any model, with cache hit rate, batch mode, and long-context tiers.

Estimate monthly spend →
Side by Side

Compare Models

Put up to three models against the same workload and see where the price gap actually comes from: input, output, or caching.

Compare models →

By model maker

Each maker's lineup, with tiers, modalities, and rates on one page.

AI21 Labs

Tel Aviv, Israel · Est. 2017

5 models · Text

Input from $0.20 / 1M tokens

Alibaba (Qwen)

Hangzhou, Zhejiang, China · Est. 1999

7 models · Multimodal, Text

Input from $0.14 / 1M tokens

Amazon

Seattle, Washington, USA · Est. 2006

31 models · Audio / Speech, Embedding, Image, Multimodal, Text, Video

Input from $0.020 / 1M tokens

Anthropic

San Francisco, California, USA · Est. 2021

30 models · Multimodal, Text

Input from $0.10 / 1M tokens

Cohere

Toronto, Ontario, Canada · Est. 2019

8 models · Embedding, Image, Text

Input from $0.10 / 1M tokens

DeepSeek

Hangzhou, Zhejiang, China · Est. 2023

3 models · Text

Input from $0.58 / 1M tokens

Google

Mountain View, California, USA · Est. 1998

6 models · Text

Input from $0.040 / 1M tokens

Luma AI

San Francisco, California, USA · Est. 2021

1 model · Video

Meta

Menlo Park, California, USA · Est. 2004

16 models · Image, Multimodal, Text

Input from $0.10 / 1M tokens

MiniMax

Shanghai, China · Est. 2021

3 models · Text

Input from $0.30 / 1M tokens

Mistral AI

Paris, France · Est. 2023

14 models · Audio / Speech, Multimodal, Text

Input from $0.040 / 1M tokens

Moonshot AI

Beijing, China · Est. 2023

3 models · Text

Input from $0.60 / 1M tokens

NVIDIA

Santa Clara, California, USA · Est. 1993

4 models · Multimodal, Text

Input from $0.060 / 1M tokens

OpenAI

San Francisco, California, USA · Est. 2015

8 models · Text

Input from $0.070 / 1M tokens

Stability AI

London, United Kingdom · Est. 2019

4 models · Image, Text

TwelveLabs

San Francisco, California, USA · Est. 2021

4 models · Embedding, Video

Writer

San Francisco, California, USA · Est. 2020

3 models · Multimodal, Text

Input from $0.15 / 1M tokens

xAI

Palo Alto, California, USA · Est. 2023

3 models · Text

Input from $1.25 / 1M tokens

Zhipu AI (GLM)

Beijing, China · Est. 2019

3 models · Text

Input from $0.070 / 1M tokens

By kind of work

Jump into the catalog already filtered.

TextChat, summarization, code, and agents.102 models →MultimodalText plus images, documents, or video as input.25 models →EmbeddingsVectors for search, RAG, and clustering.9 models →Image generationText-to-image and image editing, priced per image.10 models →Video generationText and image-to-video, priced per second.4 models →Audio & speechSpeech in, speech out.6 models →