Tevatron/OmniEmbed-v0.1
Adapter vs Qwen/Qwen2.5-Omni-7B. Tevatron/OmniEmbed-v0.1 is a LoRA adapter, not full weights: rank 16, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 32 layers (388 adapted matrices). It modifies Qwen/Qwen2.5-Omni-7B at load time.
1 derivatives checked
- Tevatron/OmniEmbed-v0.1-multivent: Adapter