Iris-pro

Name: Iris-pro
Author: Allspark

Allspark

active

Compare with other models →

Context window256K tokens

Version History

Iris-promajorSeptember 13, 2026

AllSpark released Iris-pro (397B parameters, built on Qwen3.5-397B-A17B) alongside the smaller Iris-mini (35B parameters, built on Qwen3.6-35B-A3B), both open-weight search agents trained via a reverse-engineered link-graph question pipeline and SFT-RL climbing. Iris-pro leads or ties among open-weight models in its size class on BrowseComp, BrowseComp-ZH, DeepSearchQA, and Humanity's Last Exam.

Coverage

model release

AllSpark's Iris-mini and Iris-pro Top Open-Weight Search Agent Benchmarks

Chinese lab AllSpark has released Iris-mini and Iris-pro, two open-weight search agents built on Qwen3 models that claim the top spot among open-weight systems in their size classes on four research benchmarks. The release includes model weights, an agent harness, and evaluation code, with training pipelines to follow.

September 13, 2026 · 1:06 PM4 min read

search-agents open-weight-models Qwen3