Why does GUNICORN + DJANGO crash when there is an influx of load?

Question

Django

There is a VPS 12 cores 8Gig. A chat bot for VK hangs on it (up to 100rps at the peak). There are many failures in server responses
nginx config:

spoiler

user www-data;
worker_processes 12;
pid /run/nginx.pid;
include /etc/nginx/modules-enabled/*.conf;

events {
  use epoll;
  worker_connections 1666;
  multi_accept on;
  accept_mutex off;
}

http {

  ##
  # Basic Settings
  ##



  sendfile on;
  tcp_nopush on;
  tcp_nodelay on;
  keepalive_timeout 30;
  types_hash_max_size 2048;
  # server_tokens off;

  # server_names_hash_bucket_size 64;
  # server_name_in_redirect off;

  include /etc/nginx/mime.types;
  default_type application/octet-stream;

  ##
  # SSL Settings
  ##

  #ssl_protocols TLSv1 TLSv1.1 TLSv1.2; # Dropping SSLv3, ref: POODLE
  #ssl_prefer_server_ciphers on;

  ##
  # Logging Settings
  ##

  access_log off;
  error_log /var/log/nginx/error.log;

  ##
  # Gzip Settings
  ##

  gzip on;

  gzip_vary on;
  gzip_proxied any;
  gzip_comp_level 6;
  gzip_buffers 16 8k;
  gzip_http_version 1.1;
  gzip_types text/plain text/css application/json application/javascript text/xml application/xml application/xml+rss text/javascript;

  ##
  # Virtual Host Configs
  ##

  include /etc/nginx/conf.d/*.conf;
  include /etc/nginx/sites-enabled/*;
}

# Place this file in the `/etc/nginx/conf.d/` directory and update nginx
# configurations: sudo nginx -s reload

upstream django_project {

    server unix:/home/django/projects/<hidden>.ru/world/var/run/guni.sock fail_timeout=0;
}
# конфиг прокси 
server {
    listen 80;
    server_name www.<hidden>.ru;
    return 301 http://<hidden>.ru$request_uri;
}

server {
    # DEFAULTS

    listen 80;
    keepalive_timeout 25;
    keepalive_requests 100;
    client_max_body_size 50m;
    client_body_timeout 15;
    # If the client stops reading data, free up the stale client connection after this much time.
    send_timeout 2;
    server_name <hidden>.ru *.<hidden>.ru;


    root /home/django/projects/<hidden>.ru/world/var/www;
    access_log off;
    #/home/django/projects/<hidden>.ru/world/var/log/nginx_access.log;
    # access_log /home/django/projects/<hidden>.ru/world/var/log/$host;

    error_log /var/log/nginx/error.log;

    # OPTIMIZATION
    proxy_connect_timeout 16;
    proxy_send_timeout 16;
    proxy_read_timeout 16;

    # HANDLERS
    location ~ ^/(media|static)/  {
        # Buffering the static data for 7 minutes.
        expires 7m;
    }

    location / {
        # Checks for static file, if not found proxy to app.
        try_files $uri @django_app_proxy;
    }

    location /nginx_status {
        stub_status;
    }

    location @django_app_proxy {
        # DEFAULTS
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header Host $http_host;
        proxy_buffering off;
        proxy_redirect off;

        # OPTIMIZATION
        # fastcgi_read_timeout 300;
        # limit_req zone=one burst=5;
        proxy_read_timeout 300;

        # HANDLERS
        proxy_pass http://django_project;
    }
}

I run gunicorn like this (via supervisor):
gunicorn --bind "unix:$BASE_DIR/world/var/run/guni.sock" \
--pid="$BASE_DIR/world/var/run/gunicorn.pid" \
-- log-file "$BASE_DIR/world/var/log/gunicorn_stdout.log" \
--timeout=20 \
--workers=25 \
"basic.wsgi:application"
What could be causing the crashes? Is the server weak? Configs not optimized?
The profiler shows that the code is executed in 150-200ms. (on SQL - 50ms leaves).