ggml-org/SmolVLM-500M-Instruct-GGUF

Same weights, converted (fewer layers) vs HuggingFaceTB/SmolVLM-500M-Instruct. ggml-org/SmolVLM-500M-Instruct-GGUF matches part of HuggingFaceTB/SmolVLM-500M-Instruct (291 of its 489 tensors; the rest of HuggingFaceTB/SmolVLM-500M-Instruct has no counterpart). On the shared part: Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 0/128

On the Hub · Downloads on Model Pulse