GPT-4o mini

flackz

flackz

Senior Member
Joined
Nov 28, 2023
Messages
834
Reaction score
345
The new affordable and intelligent small model that’s significantly smarter, cheaper, and just as fast as GPT-3.5 Turbo—and launched it in the API. Here’s what you need to know:

  • Intelligence: GPT-4o mini outperforms GPT-3.5 Turbo in textual intelligence (scoring 82% on MMLU compared to 69.8%) and multimodal reasoning.
  • Price: GPT-4o mini is more than 60% cheaper than GPT-3.5 Turbo, priced at $0.15 per 1M input tokens and $0.60 per 1M output tokens (roughly the equivalent of 2500 pages in a standard book).
  • Modalities: GPT-4o mini currently supports text and vision capabilities, and we plan to add support for audio and video inputs and outputs in the future.
  • Languages: GPT-4o mini has improved multilingual understanding over GPT-3.5 Turbo across a wide range of non-English languages.
 
Advertise on BHW
Good numbers, good prices. Bye bye local LLMs for a while :) (if not for censored things)
 
Cutoff date? Oct 2023 meh. For me cutoff date is more important.
 
this model just saved me 90% of the costs for many content generation tasks :D
not as good as gpt-4o or gpt-4-turbo tho, but works well for specific cases, surely better than gpt-3 and it's even cheaper,
gpt-3 will be dying for most case uses after this one
 
IM Dude said:
Cutoff date? Oct 2023 meh. For me cutoff date is more important.
Click to expand...
Check Retrieval Augmented Generation and e.g. LlamaIndex.

New thing -> Get Serps -> Get Urls -> Get Embeddings -> Vector-Search -> Augment prompt automatically with relevant information -> Up-To-Date Generation

Can be done with a small python script.
 
well, until is tested who knows if is really good and sometimes the down grade the quality when they see the success, I will believe is the same gpt 3.5 just with a different name.
 
MoonStarz said:
well, until is tested who knows if is really good and sometimes the down grade the quality when they see the success, I will believe is the same gpt 3.5 just with a different name.
Click to expand...

Here is a comparison of GPT-4o with GPT-4o mini in a real-life auto-blogging task.


I see no reason to continue using GPT-4o after that.
 
4o-mini is good for specific small tasks to replace fine-tuned 7b, 8b.

I've used them to replace some of the small agents; it works pretty well.

For tasks that require general information, use the bigger ones. 4o-mini is too small anyway.
 
4440 said:
For tasks that require general information, use the bigger ones. 4o-mini is too small anyway.
Click to expand...

The example was posted above - none of 7B, 8B and even 70B is able to produce such quality HTML-formatted texts in multiple languages.

AD_4nXfVuZJp-BgdcX-9knuhips7Jzs9pf0HNi4WeZQdhPDQyWUDC8kWrbIK7da4YVYReZRLloBG33NrA0rIuAkshd7t-DBQfC_xgmTPB_YPWlwYRlUw4_-IXaxzcUz4srqOI1SNbOJsprPt6FORNRJ6v2_rO3Y
 
This appears to be a cheap option, but we are still waiting for GPT 5.
 
Advertise on BHW
You must log in or register to reply here.
Sign up now!

Main Menu

Home Main Forum List Black Hat SEO White Hat SEO BHW Newbie Guide Blogging Black Hat Tools Social Media Downloads

Marketplace

Account Selling Content / Copywriting EBooks / Methods Hosting Misc Proxies For Sale SEO - Link building SEO - Packages Social Media Social Media - Panels Web Design More

Making Money

Affiliate Programs Hire a Freelancer IM Journeys Making Money Pay Per Click Site Flipping Want to Buy

BlackHat World

Dispute Resolution Forum Suggestions & Feedback Introductions Lounge
Back
Top