ATOM™-Max Server

대규모 AI 추론을 위한 고성능 서버

자세히 보기

ATOM™-Max Pod

RDMA 고속 네트워크 기반 Rack-scale AI 인프라

자세히 보기

Enterprise Al Solution

Full Packaged 엔터프라이즈 AI 서빙 솔루션

Enterprise Al Solution

REBEL-Quad

하이퍼스케일 멀티모달 AI·MoE 가속을 위한 차세대 NPU

자세히 보기

Rebellions

Efficient.
Scalable.
Deployment-Ready.

대규모 AI 서비스
추론 성능

대규모 요청 처리 성능과 높은 전력 효율성으로 운영 비용을 절감하고 비즈니스 효율을 극대화합니다.

300+ 모델 지원으로
바로 구현하는 AI 서비스

최신 A 모델을 폭넓게 지원하여 다양한 산업 분야의 AI 서비스를 즉시 구현할 수 있습니다.

Model Zoo

DeepSeek-R1-Distill-Llama

DeepSeek-R1-Distill-Llama is a fine-tuned model based on Meta's LLM, using samples generated by DeepSeek-R1, specialized in Chain-of-Thought reasoning.

Qwen2.5

Qwen 2.5 is an advanced language model developed by Alibaba Cloud, designed for high-performance natural language understanding and generation tasks.

DeepSeek-R1-Distill-Qwen

DeepSeek-R1-Distill-Qwen is a fine-tuned LLM based on Qwen by Alibaba Cloud, using samples generated by DeepSeek-R1, specialized in Chain-of-Thought reasoning.

Stable Diffusion

Stable Diffusion models are advanced text-to-image models that generate high-quality images from textual descriptions, with variations optimized for efficiency, fine-tuning, and creativity.

Llama3.x

Llama3 is Meta's newest language model, featuring enhanced contextual understanding and multimodal functionalities.

YOLO

YOLO (You Only Look Once) is an advanced real-time object detection model that identifies and locates multiple objects in images or videos with high accuracy and efficiency.

간편한 도입과 운영

GPU 수준의 익숙한 개발환경과 오픈소스 소프트웨어 생태계와의 호환으로 개발 시간을 단축하고 운영 효율성을 높입니다.

Developers

유연한 Scale-Up & Out 지원

다양한 Scale-Up, Scale-Out 옵션으로 투자 리스크를 최소화하고 성장에 맞춰 확장할 수 있습니다.

RSD

Our Innovation Ecosystem

Tech sovereignty starts with control over compute. Rebellions is building the kind of homegrown AI infrastructure every region will need to stay competitive.

- Fleur Pellerin, CEO, Korelya -

We see Rebellions as the most promising AI accelerator startup in APAC. Their execution speed, silicon capability, and system vision give them a real edge in a fast-moving market.

- Junichi Nakajima, Representative Director, DGDV -

Rebellions is more than a technology partner. They’re one of the few with silicon ready for real-world AI services. That’s why we’re deploying Rebellions’ NPU in A., Korea’s largest LLM service. It’s the first large-scale NPU deployment in a real AI application, and a bold step toward next-gen AI infrastructure.

- Tony Ha, Head of AI DC Business Division, SK Telecom -

Rebellions brings deep insight into real AI workloads, shaped by cutting-edge projects with global leaders. Their expertise makes them a valuable partner as we push the limits of advanced verification, including complex multi-die designs.

- Johannes Stahl, Executive Director, Product Marketing, Synopsys -

도입 사례 알아보기