---
format: "aidr-story-markdown/v1"
id: "4ceef963381f0657a12af5b9c9918e6fc99a48bb689f7efda16714e1410546b8"
canonical_url: "https://aidr.today/4ceef963?lang=en"
title: "AntLingAGI Open-Sources 560B Parameter Ling-3.1-flash Model"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-14T15:54:15.000Z"
category: "Open Source"
topics: ["open-source"]
source_urls: ["https://huggingnews.com/ai/antlingagi-open-sources-560b-parameter-ling-31-flash-model-4ec5cc4e","https://marketbrief.now/ai/antlingagi-open-sources-560b-parameter-ling-31-flash-model-4ec5cc4e"]
summary: "AntLingAGI released Ling-3.1-flash, a 560B parameter open-weights reasoning model with 25B active parameters per token and a 1M token context window, targeting coding, office, and agent tasks. The model supports vLLM and SGLang inference frameworks and uses MXFP4 and block-FP8 quantization to reduce weights from 534GB to 292GB, enabling deployment on 4 Hopper GPUs."
---

# AntLingAGI Open\-Sources 560B Parameter Ling\-3\.1\-flash Model

> [Open the canonical story](<https://aidr.today/4ceef963?lang=en>)

**Published:** 2026-10-14T15:54:15.000Z
**Category:** Open Source
**Topics:** open\-source

## Summary

AntLingAGI released Ling\-3\.1\-flash, a 560B parameter open\-weights reasoning model with 25B active parameters per token and a 1M token context window, targeting coding, office, and agent tasks\. The model supports vLLM and SGLang inference frameworks and uses MXFP4 and block\-FP8 quantization to reduce weights from 534GB to 292GB, enabling deployment on 4 Hopper GPUs\.

## Sources

- [Story source](<https://huggingnews.com/ai/antlingagi-open-sources-560b-parameter-ling-31-flash-model-4ec5cc4e>)
- [Story source](<https://marketbrief.now/ai/antlingagi-open-sources-560b-parameter-ling-31-flash-model-4ec5cc4e>)