Awesome GitHub RepositoriesBatch Request Execution

FinceptTerminal is a quantitative finance platform and financial engineering library designed for asset valuation, risk management, and fixed-income analytics. It provides a comprehensive suite for algorithmic trading and investment strategy automation, integrating specialized language model agents and node-based workflows to automate market research and alpha generation. The project distinguishes itself with a dedicated game theory analysis engine for calculating Nash equilibria and simulating strategic interactions in competitive markets. It also features a specialized credit risk modeling

Guzzle is a PHP HTTP client used for sending synchronous and asynchronous requests to web services. It serves as a concurrent HTTP request manager, an HTTP stream handler, and a middleware-based HTTP pipeline. The project is a PSR-7 compliant client, utilizing standardized PHP interfaces for requests, responses, and streams. The library differentiates itself through a customizable functional handler stack that allows for the interception and modification of the request and response lifecycle. It features an adapter-based transport system that enables swapping between network implementations,

Judge0 is an online code execution engine and multi-language compiler API designed to compile and run source code within isolated sandboxes. It functions as an asynchronous job processor that handles code submissions via a queue and provides a secure environment to run arbitrary programs while preventing unauthorized system access. The system distinguishes itself through a multi-stage compilation pipeline and a flexible execution model that supports both single-file submissions and multi-file program execution via archives. It employs an isolate-based sandboxing mechanism to enforce strict ha

This project provides the formal technical specifications and reference logic for the Ethereum proof-of-stake consensus layer. It defines the standards for block production, state transition rules, and the beacon chain logic required to ensure consistent network agreement. The implementation covers specialized mechanisms for chain security and efficiency, including fork-choice algorithms for canonical chain determination, committee-based signature aggregation, and KZG-based blob commitments for data availability. It also specifies the protocols for light client synchronization using sync comm

openai-go is an LLM SDK for Go and a client for interacting with OpenAI services. It provides type-safe bindings to generate text, images, and audio via REST endpoints, enabling the integration of large language models and AI assistant orchestration into Go applications. The library serves as an agent orchestration tool for managing stateful conversation threads and autonomous agents with integrated tool calling and file search. It also functions as an asynchronous batch processing client for monitoring large-scale request groups and fine-tuning jobs, alongside a management SDK for controllin

Această bibliotecă oferă un framework pentru impunerea constrângerilor structurale asupra output-ului modelelor de limbaj în timpul procesului de generare a token-urilor. Funcționează ca un middleware care restricționează răspunsurile modelului pentru a adera strict la scheme JSON predefinite sau la tipare de expresii regulate, asigurându-se că textul generat este lizibil pentru mașină și consistent pentru consumul ulterior de date. Proiectul se distinge prin integrarea directă cu motoarele de inferență pentru a intercepta distribuțiile de probabilitate ale token-urilor înainte de etapa finală de eșantionare. Prin utilizarea parsării cu mașini de stare și descompunerea recursivă a schemelor, efectuează validarea lookahead pentru a elimina secvențele de token-uri invalide. Această abordare permite un control precis asupra output-ului, inclusiv impunerea unei ordini specifice a câmpurilor în obiectele JSON și capacitatea de a gestiona mai multe fluxuri de generare concurente prin execuția constrângerilor în loturi (batched). Biblioteca suportă o gamă largă de strategii de integrare, permițându-i să funcționeze pe diverse backend-uri de modele și medii de server de inferență. Include instrumente de diagnosticare pentru analizarea impactului acestor constrângeri asupra performanței, asigurând compatibilitatea și eficiența pe diferite configurații hardware. Software-ul este distribuit ca un pachet Python conceput pentru integrarea în pipeline-urile de inferență existente.