Image access error!

[Sm7-git]

Bro I have attached the screenshots below, the problem is that my facebook credentials are 1000% correct i am damn sure for this.
But still the below error is show why? Even I have access to ai.meta.com website also and there also i can log with those credentials.
I have tried more than 50 times what should i do? How could i generate images ?

Before entering credentials
Input:

Output:

After Entering 100% correct credentials
Input:

Output:

[Strvm]

They seem to have added a captcha:

[mihir20]

Can we do something like, displaying the link or even captcha page to user and ask there input for captcha?

[Sm7-git]

Then does that mean its impossible to generate images now ?

[Sm7-git]

Can we do something like, displaying the link or even captcha page to user and ask there input for captcha?

No, because then what will be the need for Api if each time the user manually enters the captcha, a package or Api is made to make tasks easier by automating the task or based on programs logic. Entering manually the captcha each time is not a good idea.

[mihir20]

I thought we need to do captcha verification only once per session.

[mihir20]

For 1 of my personal projects I did this and it worked

replacing

meta-ai-api/src/meta_ai_api/main.py

Line 313 in 8965518

fb_session = get_fb_session(self.fb_email, self.fb_password)

with

fb_session = {'abra_sess':'GET_TOKEN_BY_LOGIN_TO_YOUR_BROWSER'}

And every time I want to use this library I get the token by login to meta ai. one token is valid for a few hours I guess.

[majnas]

This happens for me when running script in container, but when running in shell it's working fine

[RexsyBima]

For 1 of my personal projects I did this and it worked

replacing

meta-ai-api/src/meta_ai_api/main.py

Line 313 in 8965518

fb_session = get_fb_session(self.fb_email, self.fb_password)
with

fb_session = {'abra_sess':'GET_TOKEN_BY_LOGIN_TO_YOUR_BROWSER'}

And every time I want to use this library I get the token by login to meta ai. one token is valid for a few hours I guess.

can you guide us how exactly you get the token?? i use chrome and i dont know where its located

[rt6yr]

first go to meta.ai , make sure you are loggined there . use either meta login or facebook or instagram or create new. once login , simply click on inspect , go to network , then in prompt , tell meta to draw something like draw me a cat , switch to the network inspect one . click on any which shows as graphql , click on it , then navigate to cookies, you will see 'abra_sess' followedby a value , simply copy and paste in a textfile . now go to

meta-ai-api/src/meta_ai_api/main.py

Line 313 in 8965518

fb_session = get_fb_session(self.fb_email, self.fb_password)

, you will find fb_session = get_fb_session(self.fb_email, self.fb_password) , comment it , and then add fb_session = {'abra_sess':'abra_sess cookie , you just copied'} and save it .

note : abra_sess value is very huge , the part visible is only a part of it (about 20% of the total), so copy the entire value .

[Sm7-git]

I thought we need to do captcha verification only once per session.

Yaa, though we have to like login once per session.....but still it do not give that ease needed for api....and I think its practically impossible to verify captcha each time per session..........

And we can't trust the captcha as it's work is to stop automated things or bots and we are doing the same......maybe it askes more frequent than session.........

[siddht4]

@Sm7-git it might be true if you are you have deployed somewhere ; but for my own use / local use , I have been using the same session for past 5 months or so . the only drawback being a hard coded cookie session , still it does the work . Another trick is to use multiple or different user agent with proper header , so you are almost spoofing the API as a legitimate browser