---
format: "aidr-story-markdown/v1"
id: "13253073fa31491ad4e665dd19309ac0cb903ff016e9a1cfeac3873a2db886d6"
canonical_url: "https://aidr.today/13253073?lang=en"
title: "Making open models work for long-horizon agentic workloads"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-16T00:00:00.000Z"
category: "Tools"
topics: ["agent","devtools"]
source_urls: ["https://www.together.ai/blog/making-open-models-work-for-long-horizon-agentic-workloads"]
summary: "Together AI and Ollama explain how to protect the prefix cache from the harness to the GPU to help open models work for long-horizon agentic workloads."
---

# Making open models work for long\-horizon agentic workloads

> [Open the canonical story](<https://aidr.today/13253073?lang=en>)

**Published:** 2026-11-16T00:00:00.000Z
**Category:** Tools
**Topics:** agent, devtools

## Summary

Together AI and Ollama explain how to protect the prefix cache from the harness to the GPU to help open models work for long\-horizon agentic workloads\.

## Sources

- [Story source](<https://www.together.ai/blog/making-open-models-work-for-long-horizon-agentic-workloads>)

