stepfun-ai/Step-3.7-Flash-GGUF

Same weights, converted (fewer layers) vs stepfun-ai/Step-3.7-Flash. stepfun-ai/Step-3.7-Flash-GGUF matches part of stepfun-ai/Step-3.7-Flash (711 of its 1471 tensors; the rest of stepfun-ai/Step-3.7-Flash has no counterpart). On the shared part: Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 43 new tensors.

Tensors changed: Norms 0/181, MLP 0/261, Attention 0/225, other 0/42, lm_head 0/1, Embeddings 0/1

On the Hub · Downloads on Model Pulse