GLM 5.3 Flash Free API, Unlimited, No GPU

Author: Data Science in your pocket · Watch on YouTube

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: GLM 5.3 Flash Free API, Unlimited, No GPU · Channel: Data Science in your pocket · 03:09 · Watch on YouTube

Topics: AI developer tools, AI models, Local and open-source models, Multimodal models

Summary

The video demonstrates how to obtain and use a free API key for the GLM 5.3 Flash model through the Nvidia page. The presenter says the key is unlimited, subject to a limit of 40 requests per minute, and shows generating a key and using the provided code in Google Colab. Example questions test decimal comparison and subtraction, and the presenter notes the model’s response speed and OpenAI-compatible interface. The API can also be used from a local system.

Key points

The video focuses on getting a free API key for GLM 5.3 Flash.
The presenter describes the model as multimodal, with 320 billion total parameters and about 18 billion active parameters.
The Nvidia page offers free APIs for both GLM 5.3 and GLM 5.3 Flash.
The presenter says the API key is unlimited but limited to 40 requests per minute.
Users need to log in and generate their own API key; the presenter advises viewers not to copy the displayed key.
The example code is tested in Google Colab with questions about decimal comparison and subtraction.
The model name shown in the code is z.ai GLM 5.3 flash.
The presenter says the API is compatible with OpenAI functionalities and can be used in Colab or locally.

Chapters

Start	Chapter	Summary
00:00	Introducing GLM 5.3 Flash	The presenter introduces the free API and describes the model’s architecture and parameters.
00:31	Getting an API key	The presenter shows the Nvidia model page and explains that users must log in to generate a key with a 40-RPM limit.
01:02	Testing the API in Google Colab	The presenter copies the code snippet into Colab and tests it with questions about decimals and subtraction.
02:05	Model response and usage options	The presenter discusses latency, the model name and API compatibility, then concludes that the free API can be used in Colab or locally.

Quotes

this is an unlimited API key that you get with a limitation on the speed that is 40 RPM. — 00:31

Don’t try to copy this API key because I will be deleting it ASAP. — 01:02

The API key you need to generate for yourself it is free to create. — 02:05

This API is free to use. — 03:09

Concepts and entities

GLM-5.3-Flash, GLM-5.3, NVIDIA, Google Colab, API key, 40 RPM

Full transcript (original language)

00:00 So, hi everyone. Welcome back to Data Science in your pocket and today we will be exploring how to get the free API key for GLM 5.3 flash model. So, I think yesterday we discussed about the free API key for GLM 5.3 and now we are discussing the free API for GLM 5.3 flash. So, this is a multimodal 320 billion model which has about 18 billion active parameters. Again, a mixture of expert which uses hybrid KD and sparse MLA attention with native floating point

00:31 eight weights. Now, as you can see I’ve opened up my Nvidia page where both the model APIs are free, be it GLM 5.3 and GLM 5.3 flash. Now, how to use the model is quite easy. We just need to click on the model card. I’ll share the link in the comment section below so you can check it out. You first of all need to log in into your account. So, I have logged in from my account as you can see here. And this is an unlimited API key that you get with a limitation on the speed that is 40 RPM.

01:02 So, here’s the code snippet that you have got. As you can see here, I already generated my API key. If you’re not able to do it, just log in once and click on generate API key. This will generate the codes for you. You need to copy this code as it is. Don’t try to copy this API key because I will be deleting it ASAP. So, now we’ll be testing out this code on Google Colab as you can see here. Uh this is the code your helpful assistant and I’m asking it a question, which number is larger, 9.11 or 9.8? So,

01:33 multiple parameters that you can play around with and here I have got an output here also, 9.8 is larger. If you feel that this might be generated from some other model, let me change the prompt too. Uh what is 2 - 9? So, I’m just not I’m not trying to judge the quality of the model, but just trying here to show you how to use the API key for free. This one of the best open-source model that is released so I’ll urge you all to use the model as

02:05 well. So, I’ll hit enter. We’ll also see on the latency side. So, I think GLM 5.3 API that we used yesterday was a little slower, but this one looks fast to me. So, you just need to provide the model name as z.ai GLM 5.3 flash. The API key you need to generate for yourself it is free to create. So, don’t try to copy this. This is the base URL. And it’s compatible with OpenAI functionalities as you can see here. So, this will take a while to give you an answer. And we’ll get a chat completion message

02:36 function like this as you can see here on the content side. 0.9.8 is larger. Why? To compare decimal, align decimal places and it has given a reason also. The model looks quite good. I think it’s at par with one of the best open-source model that has been released so far. And being it flash, it’s quite small in size as compared to GLM 5.3 also. As we are talking here as you can see, our output is here. So, I think it took about a minute to give us an output. 2 - 9 = -7.

03:09 And it is able to give us a correct answer. And you can run it on from any platform be it Google Colab or from your local system. This API is free to use. So, with this it’s a wrap. I hope you try out the model. The link is in the comment section below on how to create the account. Thank you so much. I hope you have a great day.