vkryukov

ReqLLM v2 Analysis (Design Alternatives)

This gist summarizes the comparison between ReqLlmNext (metadata‑driven spike) and the current ReqLLM v1, plus a recommended v2 simplification path.

---

1) What ReqLlmNext Simplifies

• Central pipeline clarity: Resolver → Validation → Constraints → Adapter → Wire → Provider.
• Metadata-driven model expansion: adding models via LLMDB updates is genuinely faster.

vkryukov

from __future__ import annotations

import base64

from openai import OpenAI

PDF_CONTENT = b"""%PDF-1.4
1 0 obj << /Type /Catalog /Pages 2 0 R >> endobj
2 0 obj << /Type /Pages /Kids [3 0 R] /Count 1 >> endobj
3 0 obj << /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792] /Contents 4 0 R /Resources << /Font << /F1 5 0 R >> >> >> endobj

vkryukov

# Script demonstrating PDF attachment support across different LLM providers
#
# Run with: mix run scripts/pdf_attachment_issue.exs
#
# Issue: ReqLLM encodes :file content parts as `image_url` type, but some
# providers (like OpenAI) only accept image formats in image_url content parts.
#
# The encoding happens in deps/req_llm/lib/req_llm/provider/defaults.ex:651-666

alias ReqLLM.{Context, Message}

vkryukov

Image Generation Support - Architecture and Implementation Plan

Overview

This document outlines the architecture and implementation plan for adding image generation support to ReqLLM, following the library's established patterns for operation-based dispatch and provider abstraction.

Goals

1. Add first-class image generation support with generate_image/3 and generate_image!/3 functions
2. Support OpenAI (DALL-E) and Google (Gemini) as initial providers

vkryukov

# This script demonstrates the difference in behavior when using tools with
# streaming vs regular calls.

alias ReqLLM.Context

{:ok, planet_weather_tool} =
  ReqLLM.Tool.new(
    name: "planet_weather",
    description: "Gets weather on a planet",
    parameter_schema: [planet: [type: :string, required: true]],

vkryukov

---Check S2idle path S0ix Residency---:

The system OS Kernel version is:
Linux localhost-live 6.14.0-63.fc42.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Mar 24 19:53:37 UTC 2025 x86_64 GNU/Linux

---Check whether your system supports S0ix or not---:
The acpidump tool is needed to check whether low idle S0     
capability is enabled on the test platform, please install acpica-tools     
or check if the acpidump command execution failed.

vkryukov

Based on my analysis of your library, here are comprehensive improvement suggestions organized by priority and impact:

  🎯 High Priority Improvements

  1. Enhanced Error Handling & Resilience

  - Add automatic retry with exponential backoff for transient failures (429, 500, 503 errors)
  - Implement circuit breaker pattern to prevent cascading failures
  - Add configurable timeout strategies per operation type
  - Provide better error context with request details in errors

vkryukov

#!/usr/bin/env bash
###############################################################################
# codex_beam_bootstrap.sh  ─  OTP 27 + Elixir 1.18 behind Codex MITM proxy
###############################################################################
go install github.com/asdf-vm/asdf/cmd/asdf@v0.18.0
asdf plugin add erlang https://github.com/michallepicki/asdf-erlang-prebuilt-ubuntu-24.04.git || true
asdf plugin add elixir

asdf install erlang 27.3.4
asdf set -u erlang 27.3.4

vkryukov

"""
Arithmetic puzzle solver. You are given a sequence of four digits, say 1,2,3,4,
and your job is to combine them with ordinary arithmetic operations (+, -, ×, and ÷)
in any order to make a target number. E.g. 24 = 1 * 2 * 3 * 4 or 24 = (1 + 2 + 3) * 4.

Some 'hard' problems from https://blog.plover.com/math/17-puzzle.html:
    1. Given 6, 6, 5, 2, get 17.
    2. Given 3, 3, 8, 8, get 24.
"""

vkryukov

package main

import (
	"fmt"
	"log"
	"math"
	"os"
	"sort"
	"strconv"
	"strings"