ggml-org/Qwen3.8-27B-GGUF

Same weights, converted vs Qwen/Qwen3.8-27B. Same weights stored in another dtype (BF16 → BF16/F32), plus 48 new tensors.

Tensors changed: Norms 0/161, MLP 0/192, Attention 0/64, Embeddings 0/1, lm_head 0/1

2 derivatives checked

• nikitos-not-bot/qwen3.8-27b: Quantization (fewer layers)
• bakhtin/Qwen3.8-27B-ED-5bpw-GGUF: Quantization

On the Hub · Downloads on Model Pulse