Anthropic’s new Claude 3.5 Haiku AI model is 4 times more expensive than its predecessor

An image of Claude 3 artwork
(Image credit: Anthropic)

Anthropic's newest AI model is now available, but Claude 3.5 Haiku costs 4 times more than its predecessor.

The new Claude 3.5 Haiku costs $1 per million input tokens and $5 per million output, that's a huge increase compared to the 25-cent input per million and $1.25 output per million of the previous model.

Anthropic had initially claimed the new AI model would remain at the same price point as the older version, but the company took to X to say that the price would increase due to higher results than expected in benchmarks.

The tweet read, "During final testing, Haiku surpassed Claude 3 Opus, our previous flagship model, on many benchmarks — at a fraction of the cost. As a result, we’ve increased pricing for Claude 3.5 Haiku to reflect its increase in intelligence,"

It's worth noting that this new 3.5 Haiku is far more powerful than the previous 3 Haiku AI model with a more up-to-date knowledge base and the ability to write longer text answers. It may be more powerful, but it's still missing some features found in the older model such as image analysis. Anthropic has said Claude 3 Haiku will remain available for users who want to use the image processing that the new 3.5 Haiku is missing. Anthropic says image input will arrive at a later date.

Claude 3.5 Haiku

Claude 3.5 Haiku is a more advanced version of the cheaper 3 Haiku and Anthropic says it outperforms the company's previous flagship AI model, Claude 3 Opus.

In that sense 3.5 Haiku is worth the increased cost, allowing users to access a far more powerful AI at a lower cost than 3 Opus which is priced at $15 per million input tokens and $75 per million output tokens. If Anthropic's claim that 3.5 Haiku outperformed 3 Opus on benchmarks is true then the 4x price hike compared to 3 Haiku is far easier to digest.

Claude 3.5 Haiku is available now and you can request API access directly from Anthropic's website.

You Might Also Like...

John-Anthony Disotto
Senior Writer AI

John-Anthony Disotto is TechRadar's Senior Writer, AI, bringing you the latest news on, and comprehensive coverage of, tech's biggest buzzword. An expert on all things Apple, he was previously iMore's How To Editor, and has a monthly column in MacFormat. He's based in Edinburgh, Scotland, where he worked for Apple as a technician focused on iOS and iPhone repairs at the Genius Bar. John-Anthony has used the Apple ecosystem for over a decade, and is an award-winning journalist with years of experience in editorial.

Read more
PLACEHOLDER
What is Claude: It’s time to talk about this clever AI chatbot
An image of Claude 3 artwork
Claude will eventually start speaking up during your chats
Claude Extended Thinking
How Claude’s 3.7's new ‘extended' thinking compares to ChatGPT o1's reasoning
Sam Altman and a smartphone.
If Sam Altman thinks OpenAI needs to raise ChatGPT prices, I might have to cancel my Plus membership
An iPhone showing the ChatGPT logo on its screen
ChatGPT-4.5 is here for Pro users now and Plus users next week, and I can't wait to try it
DeepSeek
Deepseek’s new AI is smarter, faster, cheaper, and a real rival to OpenAI's models
Latest in Artificial Intelligence
Opera AI Tabs
Opera's new AI feature brings order to your browser tab chaos
Apple WWDC 2025 announced
3 things Apple needs to do at WWDC 2025 to save Apple Intelligence, and why I'm convinced it will
Chat GPT-generated images along with source material
ChatGPT 4o image generation is so good we will never be able to trust iPhone renders (and photos) again
Gemini on a smartphone.
Gemini 2.5 is now available for Advanced users and it seriously improves Google’s AI reasoning
Pixel Studio on an phone
Pixel Studio on the Pixel 9 now lets you generate AI images of people, and the results can be terrifying
An AI face in profile against a digital background.
The race to trillion-parameter model training in AI is on, and this company thinks it can manage it for less than $100,000
Latest in News
Buzz Lightyear Space Ranger Spin Rennovations
Disney’s giving a classic Buzz Lightyear ride a tech overhaul – here's everything you need to know
Hisense U8 series TV on wall in living room
Hisense announces 2025 mini-LED TV lineup, with screen sizes up to 100 inches – and a surprising smart TV switch
Nintendo Music teaser art
Nintendo Music expands its library with songs from Kirby and the Forgotten Land and Tetris
Opera AI Tabs
Opera's new AI feature brings order to your browser tab chaos
An image of Pro-Ject's Flatten it closed and opened
Pro-Ject’s new vinyl flattener will fix any warped LPs you inadvertently buy on Record Store Day
The iPhone 16 Pro on a grey background
iPhone 17 Pro tipped to get 8K video recording – but I want these 3 video features instead