# LongCat-2.5

> LongCat-2.5 is a native multimodal MoE model for long-horizon agentic tasks, with 1.6 trillion total parameters and approximately 48 billion activated per token.

- **Organisation:** meituan-longcat
- **Category:** Vision-Language Models, Reasoning Models
- **Parameters:** 1.78T
- **Context length:** 524,288 tokens
- **Licence:** MIT (Commercial use OK)
- **Released:** 2026-11-04
- **Hugging Face:** 2.9K downloads (30 days), 278 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q2_K | 586.91 |
| Q3_K_M | 865.37 |
| Q4_K_M | 1069.87 |
| Q5_K_M | 1252.61 |
| Q6_K | 1448.4 |
| Q8_0 | 1863.93 |
| F16 | 3495.55 |

## About

LongCat-2.5 is a native multimodal MoE model for long-horizon agentic tasks, with 1.6 trillion total parameters and approximately 48 billion activated per token. Building on LongCat-2.0, it combines visual understanding with agentic coding for complex codebase development, full-project builds, creative design, and 3D tasks. Long-sequence training, large-scale RL, multi-teacher learning, and infrastructure optimizations on domestic AI accelerators underpin these capabilities.

• Native Visual Modeling: Building on LongCat-Next, native-resolution visual features are converted into discrete tokens and jointly aligned with text in a shared semantic space.

• Long-Horizon Agentic RL: Training combines real-world engineering tasks and reproducible agent failures with efficient rollout strategies and safeguards against reward hacking, improving reliable end-to-end task execution.

• Multi-Teacher Learning: Specialized agent, reasoning, and interaction teachers guide the model through full-vocabulary KL distillation and Tree Attention, strengthening multi-turn execution, reasoning, instruction following, and safety.

• Efficient Long-Context Training: Supernode-aware parallelism, efficient sharding, and communication–computation overlap improve training efficiency, achieving 86% higher per-device throughput than the DDP baseline in reported long-sequence SFT tasks.

## Alternatives

- [moonshotai/Kimi-K3](https://selfhost.directory/models/moonshotai/kimi-k3) (2.78T, kimi-k3, ≈1700.88 GB at Q4): Kimi K3 is an open-weight, native multimodal agentic model and our most capable model to date. It is a 2.8T-parameter model built on Kimi Delta Attention (KDA) and Attention Residuals (AttnRes), with native vision capabilities and a 1-million-token context window.
- [thinkingmachines/Inkling](https://selfhost.directory/models/thinkingmachines/inkling) (952B (18.1B active), Apache 2.0, ≈567.17 GB at Q4): Inkling is a general-purpose multimodal model that accepts text, image and audio inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages.
- [moonshotai/Kimi-K3.1-Pro](https://selfhost.directory/models/moonshotai/kimi-k3.1-pro) (2.78T, kimi-k3.1-pro, ≈1700.99 GB at Q4): Kimi K3.1 Pro is an open-weight, natively multimodal agentic model, and our most capable 3T-class model to date. Post-trained from Kimi K3 with large-scale joint reinforcement learning across hundreds of downstream domains, it is the first model in the K3.1 series.
- [moonshotai/Kimi-K2.7-Code](https://selfhost.directory/models/moonshotai/kimi-k2.7-code) (1.03T (33.3B active), modified-mit, ≈632.16 GB at Q4): Kimi K2.7 Code is a coding-focused agentic model built upon Kimi K2.6. With substantial improvements on real-world long-horizon coding tasks, it strengthens end-to-end task completion across complex software engineering workflows while improving token efficiency, reducing thinking-token usage by…
- [deepseek-ai/DeepSeek-V4.1-Flash](https://selfhost.directory/models/deepseek-ai/deepseek-v4.1-flash) (763B (188B active), MIT, ≈453.59 GB at Q4): We introduce DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts (MoE) model with 552B backbone parameters and support for contexts of up to one million tokens. The model natively processes images and text, and generates text autoregressively.
- [moonshotai/Kimi-K2.6](https://selfhost.directory/models/moonshotai/kimi-k2.6) (1.03T (33.3B active), modified-mit, ≈632.16 GB at Q4): Kimi K2.6 is an open-source, native multimodal agentic model that advances practical capabilities in long-horizon coding, coding-driven design, proactive autonomous execution, and swarm-based task orchestration.

More: https://selfhost.directory/models/meituan-longcat/longcat-2.5/alternatives

## FAQ

**How much VRAM does LongCat-2.5 need?**
About 1070 GB at Q4_K_M with an 8K context. Q8_0 needs about 1864 GB and full precision about 3496 GB. Longer contexts need more memory for the KV cache.

**Can LongCat-2.5 run on a consumer GPU?**
Not comfortably on a single one: it needs about 1070 GB at Q4. Split it over several GPUs, use a large unified-memory machine, or offload part of it to system RAM.

**Can I use LongCat-2.5 commercially?**
Yes. It is released under MIT. Permissive: do almost anything as long as you keep the copyright notice.

**What is the context length of LongCat-2.5?**
524,288 tokens (512K). The full window needs far more memory than an 8K context.

## Links

- huggingface: https://huggingface.co/meituan-longcat/LongCat-2.5
- homepage: https://longcat.ai/blog/longcat-2.5/
- github: https://github.com/meituan-longcat/SGLang-FluentLLM
- paper: https://arxiv.org/abs/2603.27538

---
Source: https://selfhost.directory/models/meituan-longcat/longcat-2.5