Meituan open-sources LongCat-2.5, a 1.6-trillion-parameter agent model
What happened
Meituan's LongCat team has released the weights of LongCat-2.5, a 1.6-trillion-parameter model aimed at long-horizon autonomous tasks, particularly coding. It uses a MoE architecture that activates about 48 billion parameters per token; the API supports a 1-million-token context and outputs of up to 128,000 tokens, can understand text, images and video, and allows reasoning to be toggled on and off. Reports say the model was built without using Nvidia GPUs at any stage. The release comes a week after the model was unveiled.
Written by AI from the reports · updated 13 h ago
Timeline
Follow the reports to see the event from each side.
- The DecoderSelectedMeituan releases a 1.6-trillion-parameter agent model that never touched an Nvidia GPU
Meituan has released the weights for LongCat-2.5, a 1.6-trillion-parameter model built for AI agents and trained entirely on Chinese hardware. On Terminal Bench, it scores 88.4 points, landing just behind Claude Opus 5. The sample deployment runs on 194 Huawei Ascend chips, and Meituan makes no mention of Nvidia GPUs.
Heat of this event
Current heat 7·Peak in the comparable range 10(Nov 15 20:00)·Change over 24 hours in the comparable range –
The trend compares only the same accounts observed without a break, so its range can be narrower than the current heat count. Move the pointer or tap the chart to see each hour; with a keyboard, use the left and right arrow keys.