inclusionAI releases Ling-3.1-flash-1019 with 1M-token window and 560B parameters
inclusionAI announced Ling-3.1-flash-1019, an upgraded version of its Ling-3.1-flash model, on Reddit’s AI communities. The new model has 560 billion total parameters and 25 billion active parameters per token, with a 1 million-token context window. It shows improved reasoning and agentic capabilities, scoring 1524 Elo on AA-Briefcase (up from 1400) and 68% on AutomationBench-AA (up from 62%).…
Key points
- Ling-3.1-flash-1019 has 560B total parameters, 25B active per token, and a 1M-token context window
- Model scores 1524 Elo on AA-Briefcase (up from 1400) and 68% on AutomationBench-AA (up from 62%)
- Optimizations include speculative decoding (MTP) and Hopper decode, with Docker deployment instructions provided
The release includes post-training optimizations and support for speculative decoding (MTP) and Hopper decode optimizations. Users can deploy it via SGLang or VLLM with Docker commands provided for GPUs like H200, GB300, and others. The article also references verified recipes for deployment, tool calling, and reasoning modes. No licensing details or public training-content scope changes are mentioned beyond a linked summary.
The story so far
3 episodes →- inclusionAI releases Ling-3.1-flash-1019 with 1M-token window and 560B parametersthis story
inclusionAI/Ling-3.1-flash-1019 · Hugging Face
huggingface.co · 27 October 2026
Loading the full article…
This text was published by huggingface.co. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- OpenAI and Anthropic cut AI model prices by up to 80% amid rising competition from open-weight models · 5 src
- OpenAI feeds partner content to ChatGPT as pre-summarized priority results, analysis finds · 1 src
- OpenAI releases GPT-6.1 Terra with halved API pricing and improved computer use performance · 15 src
- Baidu releases Unlimited-OCR-2 with unlimited-horizon parsing via SR-SWA · 1 src
- vLLM Semantic Router releases 27B open multimodal decision model d3-ultra · 1 src
Comments
via GitHub Discussions