Ant Group launches Ling-3.1-flash with 560B parameters

Ant Group's InclusionAI has launched Ling-3.1-flash, a language model with 560 billion total parameters and roughly 25 billion activated per token. It targets a context window of up to one million tokens, though the current two-week free trial is capped at 256,000 tokens; open-source release is planned after the trial.

• 560B total parameters, about 25B activated per token
• Designed for a context window of up to 1M tokens
• Free two-week trial limited to 256,000 tokens
• Open-source release promised after the trial