feat: add 4th patch (eagle.py image_token_index fallback), FP4→INT4 converter, fix tiktoken.model LFS pointer, update weight index

Upload convert_mtp_fp4_to_int4.py with huggingface_hub

Upload patch_vllm.py with huggingface_hub

Upload convert_mtp_fp8_to_int4.py with huggingface_hub

Upload model.safetensors.index.json with huggingface_hub

Upload patch_vllm.py with huggingface_hub

Upload tool_declaration_ts.py with huggingface_hub

Upload tokenizer_config.json with huggingface_hub

Upload tokenization_kimi.py with huggingface_hub

Upload tiktoken.model with huggingface_hub

Upload preprocessor_config.json with huggingface_hub

Upload modeling_kimi_k25.py with huggingface_hub

Upload modeling_deepseek.py with huggingface_hub

Upload model.safetensors.index.json with huggingface_hub

Upload media_utils.py with huggingface_hub

Upload kimi_k25_vision_processing.py with huggingface_hub

Upload kimi_k25_processor.py with huggingface_hub

Upload generation_config.json with huggingface_hub

Upload configuration_kimi_k25.py with huggingface_hub

Upload configuration_deepseek.py with huggingface_hub

Upload config.json with huggingface_hub

Upload chat_template.jinja with huggingface_hub

initial commit