When was LongCat-2.5 released?

LongCat-2.5 was released on November 1, 2026 by Meituan. This is the official LongCat-2.5 release date tracked on LLM Stats.

How much does LongCat-2.5 cost?

LongCat-2.5 pricing starts at $0.30 per million input tokens, $0.01 per million cached input tokens, $1.20 per million output tokens via Meituan, the lowest price among tracked providers.

How big is LongCat-2.5?

LongCat-2.5 has 1600 billion parameters.

Who created LongCat-2.5?

LongCat-2.5 was created by Meituan.

Is LongCat-2.5 multimodal?

Yes, LongCat-2.5 is multimodal and can accept both text and images as input.

Where can I use LongCat-2.5?

LongCat-2.5 is available through 1 provider including Meituan.

Where is the LongCat-2.5 paper or technical report?

LongCat-2.5 has a paper or technical report available at https://longcat.ai/blog/longcat-2.5. Use that source for architecture, training, release and evaluation details.

What models should I compare LongCat-2.5 against?

Common LongCat-2.5 comparisons include LongCat-2.5 vs GPT-6 Astra, LongCat-2.5 vs Muse Spark 1.3, LongCat-2.5 vs Claude Opus 4.8. Compare them side by side for benchmark scores, pricing, context window, latency and provider availability.

LongCat-2.5 Benchmarks, Pricing & Context Window

Name: LongCat-2.5
Author: Meituan

LongCat-2.5: Benchmarks, Pricing & Context Window

LongCat-2.5 is a language model from Meituan, released in November 2026, with multimodal input, a 1.0M-token context window, and pricing from $0.300/M input, $0.006/M cached input, $1.20/M output.

LongCat-2.5 is Meituan's native multimodal Mixture-of-Experts model for long-horizon agentic tasks, with about 1.6T total parameters and roughly 48B activated per token. It builds on LongCat-2.0 with stronger native visual understanding

Input

TextImageVideo

Output

Text

LongCat-2.5 benchmarks

Capability tiers

Standing within each category, adjusted for leaderboard depth.

Real tasks performance

High-confidence performance for LongCat-2.5 across real-world prompt categories. Only 95% intervals at most 4 points wide are shown.

Performance by conversation depth

How LongCat-2.5 holds up as conversations get longer.

Quality Tracker

LongCat-2.5 Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

llm-stats.com - Mon Nov 02 2026

Notice missing or incorrect data?

LongCat-2.5 pricing

Providers

LongCat-2.5 starts at $0.300 per million input tokens and $1.20 per million output tokens via Meituan. Reused prompt prefixes cost $0.0060 per million cached input tokens.

Provider	Input $/M	Cached input $/M	Output $/M	Context in / out	TTFT p95 s	Output p5 c/s	Modalities in / out
Meituan	$0.300	$0.0060	$1.20	1.0M/131.1K	—	—	/

Cached input is the discounted price for prompt tokens served from a provider cache. TTFT is time to first token. Output is characters per second; p5 is the sustained floor exceeded by 95% of observed requests.

LongCat-2.5 context window

Input and output token limits for LongCat-2.5, plus how it ranks on long-context understanding.

InputOutput

1.0Mtokens

131Ktokens

≈ 1.6k pages of text

1.0M

8K128K1M

Try now

huggle

LongCat-2.5in Huggle

Make it with
LongCat-2.5.

LongCat-2.5 latency

LongCat-2.5 time to first token, sustained output throughput, and failed-request rate from live model usage over the trailing 7 days.

LongCat-2.5 examples

Recent arena outputs from LongCat-2.5, picked from the highest-ranked matchups.

LongCat-2.5 license

LongCat-2.5 has 1.6T parameters.

Parameters: 1.6T

LongCat-2.5 resources

Official sources for LongCat-2.5: provider documentation, official playground, official launch post.

LongCat-2.5 vs other models

The most-compared alternatives to LongCat-2.5 are GPT-6 Astra, Muse Spark 1.3, Claude Opus 4.8. Open any pair side-by-side for benchmarks, pricing, context, and latency.