mradermacher/aegis-qwen2vl-3b-GGUF

Same weights, converted (fewer layers) vs sanaX3065/aegis-qwen2vl-3b. mradermacher/aegis-qwen2vl-3b-GGUF matches part of sanaX3065/aegis-qwen2vl-3b (435 of its 825 tensors; the rest of sanaX3065/aegis-qwen2vl-3b has no counterpart). On the shared part: Same weights stored in another dtype (F16 → F16/F32).

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/73, MLP 0/108, Attention 0/252

On the Hub · Downloads on Model Pulse