gray117/Swift-1.5-Qwen3.8-27B-Q4_0_ROCMFP4_STRIX

Quantization vs ukisai/Swift-1.5-Qwen3.8-27b. gray117/Swift-1.5-Qwen3.8-27B-Q4_0_ROCMFP4_STRIX is a GGUF quantization (T101/Q6_K) of the same weights: 161 tensors match exactly and 53 within quantization error.

Tensors changed: Norms 0/161, MLP 0/27, Attention 0/24, Embeddings 0/1, lm_head 0/1

On the Hub · Downloads on Model Pulse