---
title: "Nanbeige releases 33B open agentic coding model"
url: https://www.parallelquant.com/posts/nanbeige-releases-33b-open-agentic-coding-model-bbde0a
source_name: "MarkTechPost"
source_url: https://www.marktechpost.com/2026/10/25/nanbeige-releases-nanbeige4-3-33b-open-agentic-coding-model/
published: 2026-10-25T21:15:21.000Z
topics: ["open source", "llms"]
publisher: "Parallel Quant"
---

# Nanbeige releases 33B open agentic coding model

*2026-10-25 · Source: [MarkTechPost](https://www.marktechpost.com/2026/10/25/nanbeige-releases-nanbeige4-3-33b-open-agentic-coding-model/)*

Nanbeige LLM Lab released Nanbeige4.3-33B, a 33B-parameter dense model under Apache 2.0 with a 512K-token context window. It reports 80.9 on Terminal-Bench 2.1. The lab says it will be used to teach the upcoming Nanbeige4.3-3B.

**Why it matters:** A permissively licensed model at this size reporting strong terminal-agent scores adds to the stream of open coding models that fit on a single workstation. Using the large model to teach a 3B sibling reflects the growing distillation pattern, aimed at pushing agent capability onto local hardware, as with other recent small open releases.

**Topics:** open source, llms

---
Read the original: https://www.marktechpost.com/2026/10/25/nanbeige-releases-nanbeige4-3-33b-open-agentic-coding-model/
Canonical: https://www.parallelquant.com/posts/nanbeige-releases-33b-open-agentic-coding-model-bbde0a
Published by Parallel Quant — https://www.parallelquant.com
