---
title: "AI Model Comparison: API Price and Context | RZ Automate"
description: "Compare AI models from OpenAI, Anthropic, Google, DeepSeek and more. API price per 1M tokens, context window, features and what a real job costs."
url: https://rzautomate.com/models
---

# Compare AI models

Every major AI model in one table, priced per 1 million tokens. See what each one costs, how much it can read at once, and what a real job like answering support emails would cost you. Prices come from each provider's own pricing page.

* [Anthropic](https://rzautomate.com/models?provider=Anthropic#models)
* [OpenAI](https://rzautomate.com/models?provider=OpenAI#models)
* [Google](https://rzautomate.com/models?provider=Google#models)
* [xAI](https://rzautomate.com/models?provider=xAI#models)
* [DeepSeek](https://rzautomate.com/models?provider=DeepSeek#models)
* [Moonshot](https://rzautomate.com/models?provider=Moonshot#models)
* [Alibaba](https://rzautomate.com/models?provider=Alibaba#models)
* [Meta](https://rzautomate.com/models?provider=Meta#models)

## AI model pricing comparison

Prices are per 1 million tokens in US dollars. Sort any column or open a model for its full specs.

 All providers

All providersAnthropicOpenAIGooglexAIDeepSeekMoonshotAlibabaMeta

All tiers

All tiersFlagshipMidSmall

Tool callingVisionReasoningOpen weights

Results updated

Output / 1MInput / 1MContextFeatures

* [Claude Haiku 5.5Anthropic · SmallOutput / 1M$0.50Input / 1M$0.10Context1M](https://rzautomate.com/models/claude-haiku-5-5)
* [GPT-6 LunaOpenAI · SmallOutput / 1M$0.50Input / 1M$0.10Context1.05M](https://rzautomate.com/models/gpt-6-luna)
* [DeepSeek V4.1 FlashDeepSeek · SmallOutput / 1M$1.20Input / 1M$0.30Context1MOpen](https://rzautomate.com/models/deepseek-v4-1-flash)
* [Gemini 3.8 FlashGoogle · MidOutput / 1M$3.75Input / 1M$0.75Context1.05M](https://rzautomate.com/models/gemini-3-8-flash)
* [DeepSeek V4 ProDeepSeek · FlagshipOutput / 1M$3.96Input / 1M$1.32Context1MOpen](https://rzautomate.com/models/deepseek-v4-pro)
* [Muse SparkMeta · MidOutput / 1M$4.25Input / 1M$1.25Context1.05M](https://rzautomate.com/models/muse-spark)
* [GPT-6.1 TerraOpenAI · MidOutput / 1M$5.00Input / 1M$1.00Context1.05M](https://rzautomate.com/models/gpt-6-1-terra)
* [Grok 4.7xAI · FlagshipOutput / 1M$6.00Input / 1M$2.00Context500K](https://rzautomate.com/models/grok-4-7)
* [Qwen3.8-MaxAlibaba · FlagshipOutput / 1M$6.00Input / 1M$2.00Context1M](https://rzautomate.com/models/qwen-3-8-max)
* [Claude Sonnet 5.5Anthropic · MidOutput / 1M$10.00Input / 1M$2.00Context1M](https://rzautomate.com/models/claude-sonnet-5-5)
* [Gemini 4 ArgonGoogle · FlagshipOutput / 1M$10.00Input / 1M$2.00Context1.05M](https://rzautomate.com/models/gemini-4-argon)
* [Gemini 4 CarbonGoogle · FlagshipOutput / 1M$10.00Input / 1M$2.00Context1.05M](https://rzautomate.com/models/gemini-4-carbon)
* [GPT-6.1 SolOpenAI · MidOutput / 1M$10.00Input / 1M$2.00Context1.05M](https://rzautomate.com/models/gpt-6-1-sol)
* [Kimi K3Moonshot · FlagshipOutput / 1M$15.00Input / 1M$3.00Context1.05MOpen](https://rzautomate.com/models/kimi-k3)
* [Claude Opus 5.5Anthropic · FlagshipOutput / 1M$20.00Input / 1M$4.00Context1M](https://rzautomate.com/models/claude-opus-5-5)
* [Claude Fable 5.2Anthropic · FlagshipOutput / 1M$40.00Input / 1M$8.00Context1M](https://rzautomate.com/models/claude-fable-5-2)
* [GPT-6 AstraOpenAI · FlagshipOutput / 1M$50.00Input / 1M$10.00Context1.05M](https://rzautomate.com/models/gpt-6-astra)

__AI model API pricing in US dollars per 1 million tokens. Use the column buttons to sort.__
| Model                                                                                      | Output / 1M                          | Input / 1M                           | Context                           | Features                            |  |
| ------------------------------------------------------------------------------------------ | ------------------------------------ | ------------------------------------ | --------------------------------- | ----------------------------------- |  |
| [Claude Haiku 5.5Anthropic · Small](https://rzautomate.com/models/claude-haiku-5-5)     | [$0.50](https://rzautomate.com/models/claude-haiku-5-5)    | [$0.10](https://rzautomate.com/models/claude-haiku-5-5)    | [1M](https://rzautomate.com/models/claude-haiku-5-5)    |                                     |  |
| [GPT-6 LunaOpenAI · Small](https://rzautomate.com/models/gpt-6-luna)                       | [$0.50](https://rzautomate.com/models/gpt-6-luna)          | [$0.10](https://rzautomate.com/models/gpt-6-luna)          | [1.05M](https://rzautomate.com/models/gpt-6-luna)       |                                     |  |
| [DeepSeek V4.1 FlashDeepSeek · Small](https://rzautomate.com/models/deepseek-v4-1-flash) | [$1.20](https://rzautomate.com/models/deepseek-v4-1-flash) | [$0.30](https://rzautomate.com/models/deepseek-v4-1-flash) | [1M](https://rzautomate.com/models/deepseek-v4-1-flash) | [Open](https://rzautomate.com/models/deepseek-v4-1-flash) |  |
| [Gemini 3.8 FlashGoogle · Mid](https://rzautomate.com/models/gemini-3-8-flash)             | [$3.75](https://rzautomate.com/models/gemini-3-8-flash)    | [$0.75](https://rzautomate.com/models/gemini-3-8-flash)    | [1.05M](https://rzautomate.com/models/gemini-3-8-flash) |                                     |  |
| [DeepSeek V4 ProDeepSeek · Flagship](https://rzautomate.com/models/deepseek-v4-pro)      | [$3.96](https://rzautomate.com/models/deepseek-v4-pro)     | [$1.32](https://rzautomate.com/models/deepseek-v4-pro)     | [1M](https://rzautomate.com/models/deepseek-v4-pro)     | [Open](https://rzautomate.com/models/deepseek-v4-pro)     |  |
| [Muse SparkMeta · Mid](https://rzautomate.com/models/muse-spark)                             | [$4.25](https://rzautomate.com/models/muse-spark)          | [$1.25](https://rzautomate.com/models/muse-spark)          | [1.05M](https://rzautomate.com/models/muse-spark)       |                                     |  |
| [GPT-6.1 TerraOpenAI · Mid](https://rzautomate.com/models/gpt-6-1-terra)                   | [$5.00](https://rzautomate.com/models/gpt-6-1-terra)       | [$1.00](https://rzautomate.com/models/gpt-6-1-terra)       | [1.05M](https://rzautomate.com/models/gpt-6-1-terra)    |                                     |  |
| [Grok 4.7xAI · Flagship](https://rzautomate.com/models/grok-4-7)                              | [$6.00](https://rzautomate.com/models/grok-4-7)            | [$2.00](https://rzautomate.com/models/grok-4-7)            | [500K](https://rzautomate.com/models/grok-4-7)          |                                     |  |
| [Qwen3.8-MaxAlibaba · Flagship](https://rzautomate.com/models/qwen-3-8-max)               | [$6.00](https://rzautomate.com/models/qwen-3-8-max)        | [$2.00](https://rzautomate.com/models/qwen-3-8-max)        | [1M](https://rzautomate.com/models/qwen-3-8-max)        |                                     |  |
| [Claude Sonnet 5.5Anthropic · Mid](https://rzautomate.com/models/claude-sonnet-5-5)     | [$10.00](https://rzautomate.com/models/claude-sonnet-5-5)  | [$2.00](https://rzautomate.com/models/claude-sonnet-5-5)   | [1M](https://rzautomate.com/models/claude-sonnet-5-5)   |                                     |  |
| [Gemini 4 ArgonGoogle · Flagship](https://rzautomate.com/models/gemini-4-argon)            | [$10.00](https://rzautomate.com/models/gemini-4-argon)     | [$2.00](https://rzautomate.com/models/gemini-4-argon)      | [1.05M](https://rzautomate.com/models/gemini-4-argon)   |                                     |  |
| [Gemini 4 CarbonGoogle · Flagship](https://rzautomate.com/models/gemini-4-carbon)          | [$10.00](https://rzautomate.com/models/gemini-4-carbon)    | [$2.00](https://rzautomate.com/models/gemini-4-carbon)     | [1.05M](https://rzautomate.com/models/gemini-4-carbon)  |                                     |  |
| [GPT-6.1 SolOpenAI · Mid](https://rzautomate.com/models/gpt-6-1-sol)                       | [$10.00](https://rzautomate.com/models/gpt-6-1-sol)        | [$2.00](https://rzautomate.com/models/gpt-6-1-sol)         | [1.05M](https://rzautomate.com/models/gpt-6-1-sol)      |                                     |  |
| [Kimi K3Moonshot · Flagship](https://rzautomate.com/models/kimi-k3)                      | [$15.00](https://rzautomate.com/models/kimi-k3)            | [$3.00](https://rzautomate.com/models/kimi-k3)             | [1.05M](https://rzautomate.com/models/kimi-k3)          | [Open](https://rzautomate.com/models/kimi-k3)             |  |
| [Claude Opus 5.5Anthropic · Flagship](https://rzautomate.com/models/claude-opus-5-5)    | [$20.00](https://rzautomate.com/models/claude-opus-5-5)    | [$4.00](https://rzautomate.com/models/claude-opus-5-5)     | [1M](https://rzautomate.com/models/claude-opus-5-5)     |                                     |  |
| [Claude Fable 5.2Anthropic · Flagship](https://rzautomate.com/models/claude-fable-5-2)  | [$40.00](https://rzautomate.com/models/claude-fable-5-2)   | [$8.00](https://rzautomate.com/models/claude-fable-5-2)    | [1M](https://rzautomate.com/models/claude-fable-5-2)    |                                     |  |
| [GPT-6 AstraOpenAI · Flagship](https://rzautomate.com/models/gpt-6-astra)                  | [$50.00](https://rzautomate.com/models/gpt-6-astra)        | [$10.00](https://rzautomate.com/models/gpt-6-astra)        | [1.05M](https://rzautomate.com/models/gpt-6-astra)      |                                     |  |

Prices checked November 14, 2026 on each provider's pricing page. Spot a wrong number? Tell us at [choaib@rzautomate.com](mailto:choaib@rzautomate.com) and we'll fix it.

Comparing automation platforms too? See [workflow automation tools](https://rzautomate.com/automation-tools).

## What a real job costs on each model

Per-token prices are hard to picture. So here's the same math for jobs we build AI agents for. Each card uses the cheapest and the priciest model in the table that can use tools.

* ### Answer 1,000 support emails a month  
2,000 tokens in and 400 out per item  
Cheapest  
$0.40  
Claude Haiku 5.5  
Priciest  
GPT-6 Astra  
$40.00  
[AI customer support agents](https://rzautomate.com/services/ai-customer-support-agents)
* ### Read 100 invoices a month  
3,000 tokens in and 600 out per item  
Cheapest  
$0.06  
Claude Haiku 5.5  
Priciest  
GPT-6 Astra  
$6.00  
[AI document processing](https://rzautomate.com/services/ai-document-processing)
* ### Score 500 new leads a month  
1,500 tokens in and 200 out per item  
Cheapest  
$0.13  
Claude Haiku 5.5  
Priciest  
GPT-6 Astra  
$12.50  
[AI lead generation & CRM automation](https://rzautomate.com/services/ai-lead-generation)

monthly = count × (input\_tokens × input\_price + output\_tokens × output\_price) / 1,000,000, standard rates, no caching. Same presets as the calculator on every model page.

The model bill is rarely the expensive part. Getting the agent to answer correctly and write to the right place in your tools is where the work is.

## Quick picks

* [Cheapest outputClaude Haiku 5.5$0.50 per 1M output tokens](https://rzautomate.com/models/claude-haiku-5-5)
* [Largest contextGPT-6 AstraReads up to 1,050,000 tokens at once](https://rzautomate.com/models/gpt-6-astra)
* [Cheapest with tool callingClaude Haiku 5.5$0.50 per 1M output tokens](https://rzautomate.com/models/claude-haiku-5-5)
* [Cheapest open weightsDeepSeek V4.1 Flash$1.20 per 1M output tokens](https://rzautomate.com/models/deepseek-v4-1-flash)

Picked automatically from the table, so they change when prices do.

## Three kinds of AI models

* ### Flagship models  
The smartest and most expensive. Use them for long, hard tasks like agents that plan many steps or read huge documents.  
Output from $3.96 to $50.00 per 1M tokens

  * [DeepSeek V4 Pro$3.96 per 1M output](https://rzautomate.com/models/deepseek-v4-pro)
  * [Grok 4.7$6.00 per 1M output](https://rzautomate.com/models/grok-4-7)
  * [Qwen3.8-Max$6.00 per 1M output](https://rzautomate.com/models/qwen-3-8-max)
  * [Gemini 4 Argon$10.00 per 1M output](https://rzautomate.com/models/gemini-4-argon)  
[See all in the table →](https://rzautomate.com/models?tier=flagship#models)
* ### Mid-tier models  
Where most business work lands. Smart enough for real tasks and cheap enough to run thousands of times.  
Output from $3.75 to $10.00 per 1M tokens

  * [Gemini 3.8 Flash$3.75 per 1M output](https://rzautomate.com/models/gemini-3-8-flash)
  * [Muse Spark$4.25 per 1M output](https://rzautomate.com/models/muse-spark)
  * [GPT-6.1 Terra$5.00 per 1M output](https://rzautomate.com/models/gpt-6-1-terra)
  * [Claude Sonnet 5.5$10.00 per 1M output](https://rzautomate.com/models/claude-sonnet-5-5)  
[See all in the table →](https://rzautomate.com/models?tier=mid#models)
* ### Small and fast models  
Built for high volume. Sorting emails, pulling fields from forms and tagging leads, at a fraction of the price.  
Output from $0.50 to $1.20 per 1M tokens

  * [Claude Haiku 5.5$0.50 per 1M output](https://rzautomate.com/models/claude-haiku-5-5)
  * [GPT-6 Luna$0.50 per 1M output](https://rzautomate.com/models/gpt-6-luna)
  * [DeepSeek V4.1 Flash$1.20 per 1M output](https://rzautomate.com/models/deepseek-v4-1-flash)

Before you pick

## How to choose an AI model for your business

The cheapest model that does the job reliably wins. Here's the order we check things in.

1. 01  
### Can it use tools?  
If the agent has to update your CRM or send an email, it needs tool calling.
2. 02  
### Is it accurate on your data?  
We test a few models on real examples from your business before we build.
3. 03  
### Does the context fit?  
Long contracts and big help centers need a bigger window. Short tasks don't.
4. 04  
### What does it cost at your volume?  
Multiply by a month of real work, not by one request.
5. 05  
### Can you switch later?  
Prices change every few months, so the model should be easy to swap.

Our take

We don't sell any model, so we pick whichever one does your job well for the least money. Often that's a mid-tier or small model, not the flagship.

[Not sure which one fits?Book a free 30-min call](https://rzautomate.com/book?service=agent-development&source=models)

Where the numbers come from

## How we keep this table current

* ### Sources  
Each provider's own pricing page, checked against public trackers.
* ### Updates  
We check prices by hand whenever we update the table. The date sits under the table.
* ### What's in  
Models a business would actually use in production.
* ### No guessing  
If a number isn't confirmed, the cell shows a dash.

## AI model pricing FAQ

[See all questions →](https://rzautomate.com/faq)

### What does per 1M tokens mean?

A token is a small chunk of text, roughly three quarters of an English word. AI providers charge per million tokens, with one rate for the text you send and another for the text the model writes back. A short support email and its reply usually come to a couple of thousand tokens.

### Why does output cost more than input?

Writing text takes more computing power than reading it, so most providers charge several times more for output. If your workflow writes long answers, the output price drives most of the bill. That's why our table sorts by output price by default.

### What is the cheapest AI model API right now?

As of November 14, 2026, the cheapest model in our table by output price is Claude Haiku 5.5 at $0.50 per 1M output tokens. The cheapest one that can also use tools is Claude Haiku 5.5\. Cheapest isn't always best, though. A model that gets answers wrong costs more in fixes than it saves.

### What is a context window?

The context window is the most text a model can read in one request, counted in tokens. It includes your instructions, any files you attach and the conversation so far. A bigger window lets the model read long documents in one go, but filling it costs more per request.

### What's the difference between open-weights and closed models?

You can download open-weights models and run them on your own servers or through any hosting company. Closed models only work through their maker's API. Open weights give you more control over data and cost. Closed models are usually easier to start with.

### How often are these prices updated?

We check the table by hand whenever we update it, and the date under the table shows the last check. Providers change prices with little notice, so confirm on the provider's site before you set a budget. If you spot a wrong number, email us and we'll fix it.

## Picking a model is the easy part

The hard part is getting it to do real work inside your tools. We build AI agents and automations that let your business take on more work without adding headcount. On the call we'll tell you which model fits your task and what it would cost at your volume.

* Email  
[choaib@rzautomate.com](mailto:choaib@rzautomate.com)
* Reply time  
Within two hours
* 160automations running in production
* 32businesses served
* 2,000 hrsof manual work taken off clients every month

Or book directly at [cal.com/rzautomate/30min](https://cal.com/rzautomate/30min)

Want the full picture? See [AI agent development](https://rzautomate.com/services/ai-agent-development).
