GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Junhauwong
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
Junhauwong
Model Tree
Input Modalities
Output Modalities
Supported Functionality
The Apex of Local Intelligence.
Surge-Cognition-8x70B is a massive-scale Mixture of Experts (MoE) designed for users who require zero-compromise reasoning, creativity, and instruction adherence.
By utilizing a high-fidelity 8-expert architecture, Surge-Cognition activates only the most relevant pathways for any given task, providing the depth of a half-trillion parameter model with optimized efficiency.
This model is distributed in raw Safetensors format. For consumer hardware, quantization (GGUF/EXL2) is highly recommended.