Qwen 3.8 27B available on Cerebras at 1500 tok/SEC

Available Models Model Compression Frequently Asked Questions Will you change a model's architecture without notice? Where can I find your REAP pruned models? What are compression, quantization, and pruning? Quantization: Reducing the precision of numbers used to represent model weights (e.g., converting from FP16 to FP8). This reduces memory usage without changing the model’s architecture. Pruning: Permanently removing parts of a model, like layers or experts, to reduce model size. This changes the model’s architecture and creates a different model.