@MaximeRivest: I have just realesed lm15 1.0 on pypi. I hope it will serve you as well as it is serving me. lm15 is going straight to …
Summary
The post announces the release of lm15 1.0 on PyPI, a lightweight, zero-dependency Python SDK for provider-agnostic LLM calls, which is open-source and MIT licensed.
View Cached Full Text
Cached at: 09/26/26, 01:02 PM
I have just realesed lm15 1.0 on pypi. I hope it will serve you as well as it is serving me.
lm15 is going straight to the network layer, in python, it has 0 dependencies. As such, it is smaller and faster then any other sdks.
If you want a, liteweight, provider agnostic way to call llms in a universal and typed way. You should give it a try.
It was developped specifically as a delighful lower level abstraction for other AI framework to build on top.
It open sourced, mit licensed, already depended on by other AI framework (dspy).
https://lm15.dev/llms.txt pip install lm15 npm install @lm15/lm15 cargo add lm15 go get http://github.com/lm15-dev/lm15-go…
LM15 already supports over 10 providers, including typesafe.
It supports continuation, live, image gen, image inputs, audio, citation, tools, everything.
Source: https://lm15.dev/llms.txt # LM15 > LM15 is a family of model API clients built around a shared request, response, and streaming contract. Applications keep control of tool execution, retries, and conversation history. The contract defines shared behavior; the SDKs implement it using each language’s conventions. Support depends on the SDK version, provider endpoint, and model. A shared field does not guarantee universal support. ## Getting started - [Overview](https://lm15.dev/docs/): The contract, request fields, and capabilities. - [First request](https://lm15.dev/docs/first-request/): Setup links and a selectable code example. - [LM15 skill](https://lm15.dev/skill.md): Instructions for coding agents using LM15. - [Playground](https://lm15.dev/playground/): Try provider calls in the browser and copy code. Calls require credentials and may cost money. ## Authoritative sources - [Contract](https://github.com/lm15-dev/lm15-contract): Types, field rules, provider mappings, and shared test cases. - [Source precedence](https://github.com/lm15-dev/lm15-contract/blob/main/AUTHORITY.md): Which source wins when documentation, tests, and implementations disagree. - [Scope and stability](https://github.com/lm15-dev/lm15-contract/blob/main/spec/SCOPE.md): Frozen and provisional features. - [API conventions](https://github.com/lm15-dev/lm15-contract/blob/main/playbooks/api-family.md): Shared names and language-specific usage. ## Implementations - [Python](https://github.com/lm15-dev/lm15-python): Reference implementation, not the authority over the contract. - [TypeScript](https://github.com/lm15-dev/lm15-ts) - [Rust](https://github.com/lm15-dev/lm15-rs) - [Go](https://github.com/lm15-dev/lm15-go) - [Julia](https://github.com/lm15-dev/lm15-jl) Check each repository for installation, release status, and supported features. Some website guides are still under construction; use the contract and the selected SDK’s source for details.
Similar Articles
@MaximeRivest: LM15-python, a specialized library to make llm sync and async calls to any providers under a universal api model. Is go…
LM15-python is a lightweight Python library that provides a universal API for making synchronous and asynchronous calls to various LLM providers.
@MaximeRivest: litellm was sincerely a wonderful inspiration for lm15. My drive to make lm15 came from using litellm, so it was useful…
The tweet explains how litellm inspired the creation of lm15, a new tool intended to be a standard LLM interface that unifies various models.
llm 0.32a1
Release announcement for llm version 0.32a1, a Python CLI and library for working with Large Language Models, by Simon Willison.
@MaximeRivest: DSPy 3.4 release candidate it out, we are moving off litellm. It always bothered me how slow dspy was to import and how…
The DSPy 3.4 release candidate is announced, moving away from litellm to improve import speed and reduce dependencies. Community members like Drew Breunig have praised the improvements.
@dbreunig: Big release with RLM improvements, optimization chaining, the start of LiteLLM decoupling, and 24 first-time contributo…
Major open-source release featuring RLM improvements, optimization chaining, and initial LiteLLM decoupling with 24 new contributors.