SergiioB/Qwen3.8-27B-GPTQ-Int4-sym-G128-MTP-BF16
Recommended-sampling numbers (Qwen model card presets) + records
Disclose temperature 0 (greedy) on all measured numbers
Card: real-measurements provenance line, dedicated cookbook link section
Card refresh: draft-INT4 overlay + concurrent serving results, cookbook & LocalMaxxing references, badges
Upload README.md with huggingface_hub
Upload video_preprocessor_config.json with huggingface_hub
Upload preprocessor_config.json with huggingface_hub
Upload README.md with huggingface_hub
Upload tokenizer_config.json with huggingface_hub
Upload tokenizer.json with huggingface_hub
Upload quantize_config.json with huggingface_hub
Upload quant_log.csv with huggingface_hub
Upload processor_config.json with huggingface_hub
Upload model.safetensors.index.json with huggingface_hub
Upload model-00005-of-00005.safetensors with huggingface_hub
Upload model-00004-of-00005.safetensors with huggingface_hub
Upload model-00003-of-00005.safetensors with huggingface_hub
Upload model-00002-of-00005.safetensors with huggingface_hub
Upload model-00001-of-00005.safetensors with huggingface_hub
Upload generation_config.json with huggingface_hub
Upload config.json with huggingface_hub
Upload chat_template.jinja with huggingface_hub
Upload README.md with huggingface_hub
initial commit