Blackbox supercharges coding agents with SambaCloud

blackbox-logo-300x300

Blackbox's collaboration with SambaNova means developers get tools that work as fast as they think.

The Challenge:
Scaling smarter, faster agents

Blackbox.ai is redefining coding with autonomous agents that empower developers to build custom applications effortlessly. Their flagship agent, CyberCoder, helps developers edit multiple files, create applications from scratch, and auto-document code—all in real-time. But as Blackbox's user base skyrocketed to over 10 million monthly active users, including Fortune 500 companies, they faced a new challenge: How to keep up with the growing performance demands for faster, smarter coding experiences?

The Solution:
Low latency, high performance with SambaNova

Enter SambaCloud. Known for delivering lightning-fast inference across open-source models like Llama, DeepSeek, and Qwen, SambaCloud became the performance engine Blackbox needed.

"We were drawn to SambaNova after seeing Andrew Ng’s post about their cloud platform," Rizk explained. "We’re using Llama models from 8 B to 405 B for everything from simple edits to complex coding challenges. The low latency and high performance are game-changers."

3-4X

Faster model completions

100s

Requests per second 

>10M

Monthly active users

“We’ve experimented extensively with Meta's Llama models since their release. They’re great for diverse use cases, and we fine-tune them to excel at specific tasks. But scaling that performance efficiently became a hurdle."


— Robert Rizk, Blackbox.ai Co-founder and CEO

BlackBox CyberCoder powered by SambaNova

 

Related resources

SambaNova Expands Deployment with SoftBank Corp. to Offer Fast AI Inference Across APAC

SambaNova Expands Deployment with SoftBank Corp. to Offer Fast AI Inference Across APAC

March 5, 2025
Qwen3 Is Here - Now Live on SambaNova Cloud

Qwen3 Is Here - Now Live on SambaNova Cloud

May 2, 2025
SambaNova Partners with Meta to Deliver Lightning Fast Inference on Llama 4

SambaNova Partners with Meta to Deliver Lightning Fast Inference on Llama 4

April 7, 2025
Let's Go!