Tag
LiteLLM 的发布管道于 2026 年 10 月 21 日被入侵，被篡改的 v1.107.0 版本向 PyPI 和 Docker Hub 推送了窃取凭据并支持远程代码执行的恶意软件，目前事件已被控制，受影响制品已下架。
The LiteLLM 1.107.0 release on PyPI was compromised with a credential stealer and an RCE backdoor installed via all platform wheels, activated on every Python startup through cook.pth and sitecustomize.py. SafeDep's analysis shows the malicious sdist did not match its GitHub tag and was published outside the public release pipeline; PyPI has since removed the release.
The tweet explains how litellm inspired the creation of lm15, a new tool intended to be a standard LLM interface that unifies various models.
A massive supply-chain attack on the open-source AI tool LiteLLM exposed terabytes of credentials from thousands of organizations, including Microsoft, Amazon, and Cisco, during a 40-minute window in March. Security firms CloudSEK and Hudson Rock disclosed the breach, attributing it to the TeamPCP gang.
The author shares experience building a human-in-the-loop approval system for an enterprise agent platform, emphasizing that the approval step must be a true blocking pause with editable parameters and first-class rejection/editing outcomes, and asks how others structure agent suspension.
LiteLLM has migrated from Python to Rust, achieving massive performance improvements: request overhead reduced by 150x to 0.05ms, throughput increased by 15x, memory usage reduced by 11x to 32MB.
An Android phone is repurposed as a portable GGUF inference server with Vulkan acceleration, exposing an OpenAI-compatible endpoint via LiteLLM and Tailscale mesh for integration into a self-hosted AI cluster.
A detailed comparison of OpenRouter, Concentrate.ai, Portkey, and LiteLLM as LLM gateways for a B2B SaaS product, covering features, costs, and governance.
kg-gen is an open-source Python package that uses language models (via LiteLLM and DSPy) to extract knowledge graphs from plain text or conversation messages, featuring chunking, clustering, and flexible provider routing.
A founder seeks validation for a service that configures production-grade LLM gateways to address common enterprise issues like cost visibility, provider lock-in, and PII leakage using open-source tools.
Major open-source release featuring RLM improvements, optimization chaining, and initial LiteLLM decoupling with 24 new contributors.
GoModel is a new open-source AI gateway written in Go that offers a unified OpenAI-compatible API for multiple providers and claims to be 44× lighter than LiteLLM.