A high-performance inference engine for LLMs, optimized for diverse AI accelerators.

xLLM 1.0, smart crawling, knowledge graph discovery.

🦖 X—LLM: Cutting Edge & Easy LLM Finetuning

A flexible serving framework that delivers efficient and fault-tolerant LLM inference for clustered deployments.

xLLM 2.0. No Blackbox, secure, accurate, auditable Enterprise AI, developed by BondingAI. Built from scratch. With proprietary agents.

[XLLM@ACL2025] Official Code for "Less is More: Enhancing Structured Multi-Agent Reasoning via Quality-Guided Distillation"

🦖 X—LLM: Simple & Cutting Edge LLM Finetuning

Demo project using XLLM

This repository features an example of how to utilize the xllm library. Included is a solution for a common type of assessment given to LLM engineers

No description available

[NAACL'25 SRW & XLLM@ACL'25] Typed-RAG: Type-Aware Decomposition of Non-Factoid Questions for Retrieval-Augmented Generation

Auto agent pipeline for xLLM.

A lightweight llama2 inference framework. It can inference llama2-7b with 166+ tokens/s on signle 4090.

xllmapi - 一个大模型共享网络

A lightweight llama2 inference framework. It can inference llama2-7b with 166+ tokens/s on signle 4090.

No description available

Xenium LLMNR Responder – a Link-Local Multicast Name Resolution (LLMNR) responder daemon.

Code for the BARTABSA++ Paper presented at the XLLM@ACL 2025 workshop.

Demo project using XLLM

No description available

AI acceleration by profiling models

No description available

Any LLM Perf Test

A desktop GUI for querying OpenAI models interactively or in bulk from CSV files. Supports GPT-4, GPT-4o, and GPT-3.5-turbo with secure API handling, real-time chat, and automated CSV output generation.

No description available

User-Defined Trade-Offs in LLM Benchmarking: Balancing Accuracy, Scale, and Sustainability

Explainable LLM

No description available

No description available

ACL 2024 | LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback