LongCat-2.5-Preview is now live. 1.6T parameters. ~48B active. A 1M-token context window. Natively multimodal. Built to take on long-horizon tasks. From terminals and browsers to GUIs, spreadsheets, and design tools. Try it now: 🚀 API: longcat.ai/platform/ 💬 Chat: longcat.ai/chat/

Sep 25, 2026 · 2:16 PM UTC

74
124
1,346
155,751
🎁 To everyone who has supported us: 5M free tokens. Every existing user gets 5 million free tokens to try LongCat-2.5-Preview. Check your balance on longcat.ai/platform/usage. Token packages: all existing token packages work with both LongCat-2.5-Preview and LongCat-2.0. Any questions or feedback, join our discord for support: discord.gg/jmuv5TQzKU
7
6
103
12,230
Sort replies: Relevant Recent Liked
Replying to @Meituan_LongCat
benchmark ,你在哪...
4
46
3,936
Replying to @Meituan_LongCat
In honour of LongCat i present to you my cats
2
33
2,018
Replying to @Meituan_LongCat
Thanks for sharing the benchmarks!
28
3,033
Replying to @Meituan_LongCat
怎么没有 benchmark 评分情况?
3
11
4,290
Replying to @Meituan_LongCat
the food delivery app just pulled up with a trillion parameter side quest. 😂
6
1,728
Replying to @Meituan_LongCat
Can we have the benchmarks please?
1
6
1,553
Replying to @Meituan_LongCat
这种类型的公告帖,应该附加benchmark成绩/竞品模型对比,不然都没人知道模型的定位,更加不会有人直接买API的啊😅 要是再好点,可以上OpenRouter之类搞一个短期免费试用的,更有说服力。 longcat你们都做到2.5了,发公告应该能熟练点了吧
1
3
621
Replying to @Meituan_LongCat
@grok what's the pricing
1
1
6,046
Replying to @Meituan_LongCat
能否加上一些简单的性能与特性介绍?
1
2
3,093
Replying to @Meituan_LongCat
is it actually 1M context or does it become unusable at 500k tokens?
2
287
Replying to @Meituan_LongCat
Where are the weights??

ALT Fast And Furious GIF by The Fast Saga

3
1,199
Replying to @Meituan_LongCat
Benchmarks?
2
1
1,812
Replying to @Meituan_LongCat
Is it any faster than 2.0?
1
2,626
Replying to @Meituan_LongCat
连跑分都没有?
1
1
2,349
Replying to @Meituan_LongCat
1.6T 预览又来了。上下文百万,账单也跟着长。
855
Replying to @Meituan_LongCat
I will facing log in issue from Bangladesh ..
1
1,768
Replying to @Meituan_LongCat
不知道效果如何,有评论家来评评吗?
805
Replying to @Meituan_LongCat
thank u! good job
1,724
Replying to @Meituan_LongCat
性能介绍一下呀。
438
Replying to @Meituan_LongCat
could you define what a "long-horizon task" is?
41
Replying to @Meituan_LongCat
@grok whats the price on it? And are there any Benchmarks?
1
2,316
Replying to @Meituan_LongCat
ready to build now
2,011
Replying to @Meituan_LongCat
Awesome!gonna try it out
2,202
Replying to @Meituan_LongCat
Big fan of 2.0 Let's see how good 2.5 is
1
1,003
Replying to @Meituan_LongCat
Why tf do you need by phone number just to visit the website
281
Replying to @Meituan_LongCat
This model loops as hell tfff
1
42
Replying to @Meituan_LongCat
holy crap meituan is cooking
561
Replying to @Meituan_LongCat
Awesome!gonna try it out
1,337
Replying to @Meituan_LongCat
Any bench ?
866
Replying to @Meituan_LongCat
a TRILLION parameter model by a FOOD DELIVERY startup btw. maybe a dumb question, but how do these seemingly random Chinese companies have the resources/talent to make their own models?
4
21
1,987
Replying to @Meituan_LongCat
Meituan is much more than a food delivery company. It’s one of China’s largest local-services platforms, with deep reach into restaurants, hotels, travel, mobility, merchants, and everyday consumer services. So it’s not that surprising to see them build a large model. China has a deep AI talent pool, and domestic compute is also becoming much more capable. LongCat-2.0 was already trained and served on a 50,000-chip Chinese accelerator cluster, so the infrastructure side is clearly catching up too.
4
945
Replying to @Meituan_LongCat
our tracker saw longcat 2.5 preview on your api 2 hours before this post
LongCat 2.5 preview just showed up in meituan's longcat api model list, two months after 2.0 was leading openrouter!!
3
1,473
Replying to @Meituan_LongCat
@grok what's the benchmarks for this and can you create a image showing them compare to gpt-6-sol astra and opus 5.5 and show some cyber benchmarks in their as well if you can find it for this model thanks
2
3
377