---
title: "AllSpark releases Iris open search agents, strongest in their class"
url: https://www.parallelquant.com/posts/allspark-releases-iris-open-search-agents-strongest-in-their-class-ef9331
source_name: "The Decoder"
source_url: https://the-decoder.com/iris-mini-and-iris-pro-are-the-strongest-open-weight-search-agents-in-their-class/
published: 2026-09-13T12:58:18.000Z
topics: ["open source", "agents", "research"]
publisher: "Parallel Quant"
---

# AllSpark releases Iris open search agents, strongest in their class

*2026-09-13 · Source: [The Decoder](https://the-decoder.com/iris-mini-and-iris-pro-are-the-strongest-open-weight-search-agents-in-their-class/)*

AllSpark released Iris-mini and Iris-pro, open-source search agents built on Qwen models. The team says the models lead benchmarks among open-weight models of similar size, and that training also improved performance on unrelated tasks like general tool use and office work.

**Why it matters:** Generalization beyond the trained task is the harder problem in agent training, so this transfer effect, if it holds up under independent testing, suggests search-agent training data has broader value than expected. It adds to the fast-growing field of open-weight agentic models competing with closed tool-use offerings from labs like OpenAI and Anthropic.

**Topics:** open source, agents, research

---
Read the original: https://the-decoder.com/iris-mini-and-iris-pro-are-the-strongest-open-weight-search-agents-in-their-class/
Canonical: https://www.parallelquant.com/posts/allspark-releases-iris-open-search-agents-strongest-in-their-class-ef9331
Published by Parallel Quant — https://www.parallelquant.com