DeepBeepMeep/Ideogram4

Add shared Qwen3-VL-8B BF16 vision tower for Qwen Image 2.1 GGUF encoders

Remove migration utility from encoder assets

Consolidate encoder assets with generic legacy export suffixes

Update README.md

Create README.md

Share encoder-only BF16 and validated INT8 ConvRot weights; replace Quanto variants

Add Ideogram4 TurboTime LoRA

Upload Qwen3-VL-8B-Instruct/chat_template.jinja

Upload Qwen3-VL-8B-Instruct/tokenizer_config.json

Upload Qwen3-VL-8B-Instruct/tokenizer.json

Upload Qwen3-VL-8B-Instruct/config.json

Upload Qwen3-VL-8B-Instruct/Qwen3-VL-8B-Instruct_nf4.safetensors

Upload Qwen3-VL-8B-Instruct/Qwen3-VL-8B-Instruct_fp8.safetensors

Upload ideogram4/ideogram4_unconditional_transformer_nf4_config.json

Upload ideogram4/ideogram4_transformer_nf4_config.json

Upload ideogram4/ideogram4_unconditional_transformer_config.json

Upload ideogram4/ideogram4_transformer_config.json

Upload Ideogram4_Unconditional_Transformer_nf4.safetensors

Upload Ideogram4_Transformer_nf4.safetensors

Upload Ideogram4_Unconditional_Transformer_fp8.safetensors

Upload Ideogram4_Transformer_fp8.safetensors

initial commit