sssemil

Cloudflare Workers Cost for an OpenRouter-Style Streaming Proxy

Pricing checked: 2026-07-26
Unit: 1,000,000 output tokens

Assumptions

Variable	Lower case	Upper case
Output tokens per request	1,000	100

sssemil

// ==UserScript==
// @name         ChatGPT Chat Exporter
// @namespace    https://gist.github.com/sssemil/b0b8af1a850aeb99b690a80dbead84c0
// @version      1.2.1
// @description  Export a ChatGPT conversation as Markdown, a print-ready PDF, or a ZIP with session files.
// @author       sssemil
// @homepageURL  https://gist.github.com/sssemil/b0b8af1a850aeb99b690a80dbead84c0
// @updateURL    https://gist.githubusercontent.com/sssemil/b0b8af1a850aeb99b690a80dbead84c0/raw/chatgpt-chat-exporter.user.js
// @downloadURL  https://gist.githubusercontent.com/sssemil/b0b8af1a850aeb99b690a80dbead84c0/raw/chatgpt-chat-exporter.user.js
// @match        https://chatgpt.com/*

sssemil

Would you use a backend where you just define schema, access policy, and functions?

Basically something like making smart contracts on EVM, but instead they run on a hyperscaler, and have regular backend fundamentals.

Here's a mock frenchie made me, was thinking something like this:

schema User {
  email: string @private(owner)
  name: string @public

sssemil

#!/usr/bin/env python3
"""
Simple CLI for PersonaPlex - talk and listen via terminal.
"""

import argparse
import os
import sys
import time
import threading

sssemil

// ==UserScript==
// @name         Auto-Authorize Claude OAuth (claude.ai)
// @match        https://claude.ai/oauth/authorize*
// @run-at       document-end
// ==/UserScript==
(() => {
  const clickAuthorize = () => {
    const btn = [...document.querySelectorAll("button")]
      .find(b => /authorize/i.test(b.textContent || ""));
    if (btn) btn.click();

sssemil

#!/usr/bin/env bash
set -e

usage() {
  echo "usage: $0 <command> -a <writable_path> [-a <writable_path>...]"
  exit 1
}

if [ "$#" -lt 3 ]; then
  usage

sssemil

Summary

txt2img sdxl 1024x1024

65.417351÷65.623950=0.996851774

img2img sdxl 1024x1024 (input 1024x1024 image)

54.623950÷55.738662=0.980001099

sssemil

import numpy as np
import torch
import cv2
from PIL import Image
from transformers import AutoImageProcessor, AutoModelForDepthEstimation
from transformers import DPTForDepthEstimation, DPTImageProcessor


def depth_estimation(model, feature_extractor, image):
    inputs = feature_extractor(images=image, return_tensors="pt").to("cuda")

sssemil

from math import inf
import torch
from torch import tensor, device
import torch.fx as fx
import torch._dynamo
from torch._dynamo.testing import rand_strided
from torch._dynamo.debug_utils import run_fwd_maybe_bwd

import torch._dynamo.config

sssemil

#!/bin/bash

set -e

# Check if zsh is installed
if ! command -v zsh &> /dev/null; then
    echo "zsh is not installed. Please install zsh first."
    exit 1
fi