mihailgribov/typecastlm-qwen3.5-3.8b

Same weights, converted (fewer layers) vs Qwen/Qwen3.5-4B. mihailgribov/typecastlm-qwen3.5-3.8b matches part of Qwen/Qwen3.5-4B (373 of its 738 tensors; the rest of Qwen/Qwen3.5-4B has no counterpart). On the shared part: Same weights stored in another dtype (BF16/F32 → BF16), plus 1 new tensors.

Tensors changed: Embeddings 0/1, MLP 0/84, Attention 0/196, Norms 0/92

On the Hub · Downloads on Model Pulse