# LMQL - new paradigm for prompting

**URL:** <https://community.openai.com/t/lmql-new-paradigm-for-prompting/297942>\
**Category:** Prompting\
**Tags:** gpt-4, prompts-as-code\
**Created:** [July 13, 2023, 3:09pm UTC](https://community.openai.com/t/lmql-new-paradigm-for-prompting/297942 "2023-07-13T15:09:15Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![bruce.dambrosio](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/bruce.dambrosio/32/224818_2.png) [@bruce.dambrosio](https://community.openai.com/u/bruce.dambrosio)\
**Post date:** [July 13, 2023, 3:09pm UTC](https://community.openai.com/t/lmql-new-paradigm-for-prompting/297942/1 "2023-07-13T15:09:15Z")

</div>

> **[2212.06094.pdf](https://arxiv.org/pdf/2212.06094.pdf)**
>
> 1553.98 KB

Based on this, we present the novel idea of Language Model Programming (LMP). LMP generalizes language  
model prompting from pure text prompts to an intuitive combination of text prompting and scripting. Additionally, LMP allows constraints to be specified over the language model output. This enables easy adaption to  
many tasks while abstracting language model internals and providing high-level semantics

---

<div class="post-metadata">

**Author:** ![jochenschultz](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/jochenschultz/32/772274_2.png) [@jochenschultz](https://community.openai.com/u/jochenschultz)\
**Post date:** [September 14, 2023, 11:17pm UTC](https://community.openai.com/t/lmql-new-paradigm-for-prompting/297942/2 "2023-09-14T23:17:08Z")

</div>

How did I miss this raw diamond of a pdf. Thank you for posting that.
