Zyphra logoZY

About

Zyphra is a full stack AGI company based in San Francisco, building the complete infrastructure for open-source superintelligence. The company develops foundation models across multiple modalities, including ZAYA for frontier reasoning language models, ZONOS for expressive text-to-speech, and ZUNA for brain-computer interface applications. Zyphra focuses on making advanced AI capabilities accessible through open models while maintaining human alignment and safety at the core of their research methodology.

The company combines deep research with engineering excellence, working across language, audio, vision, and neural signal processing to create general-purpose AI agents. Their approach emphasizes inference efficiency, novel neural architectures, and hardware-optimized performance across multiple accelerator platforms including AMD, NVIDIA, and cloud infrastructure. Zyphra collaborates with major technology companies like IBM and AMD to advance AI capabilities and push the boundaries of what artificial intelligence can achieve.

Similar companies

Together AI logoTA

Together AI

Together AI is a research-driven AI cloud infrastructure provider enabling developers and enterprises to train, fine-tune, and deploy open-source generative AI models at scale.

Sciforium logoSC

Sciforium

Sciforium builds AI infrastructure, including multimodal foundation models and a high-efficiency serving platform for rapid model deployment.

OpenAI logoOP

OpenAI

OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity.

Soris logoSO

Soris

Soris is an early-stage company building a full-stack, vertically integrated AI platform from the ground up to target large, entrenched industries like aerospace and mobility.

ME

Metamorphic

Metamorphic builds foundation models trained on neural data by combining machine learning and large-scale neuroscience to develop new approaches to intelligence.

Anthropic logoAN

Anthropic

Anthropic is an AI safety and research company building reliable, interpretable, and steerable AI systems, including the Claude family of large language models.