Any-to-Any
Transformers
Safetensors
GGUF
gemma4
multimodal
text-generation
image-text-to-text
audio-text-to-text
video-text-to-text
long-context
lora
bf16
llama-cpp
zh
en
Instructions to use chirs345678/Anrn-12B-R1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use chirs345678/Anrn-12B-R1 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("chirs345678/Anrn-12B-R1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 1,191 Bytes
4e6526a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 | {
"variant": "full-bf16-original",
"alias": "anru-gemma4-12b-native-multimodal-bf16-lora045",
"max_gpu_memory": "14GiB",
"composition": "byte-identical original BF16 base plus LoRA at scale 0.45",
"attention_implementation": "sdpa",
"default_system_prompt": "system_prompt.md",
"system_prompt_source_name": "Anru_Constitution_v2_30000.md",
"system_prompt_sha256": "5591AC2C99FDC6C8E21D1D3B87625BB368204548BEC7B63C3FD6598CF23639BF",
"system_prompt_bytes": 105352,
"system_prompt_chars": 38029,
"system_prompt_lines": 1984,
"llama_service": {
"launcher": "Start-Llama-Service.bat",
"bootstrap": "llama-bootstrap.ps1",
"runtime": "bundled when packaged; downloads latest official llama.cpp Windows build if missing",
"gguf_model": "llama-model/anru-gemma4-12b-base-Q5_K_M.gguf",
"gguf_lora": "llama-model/anru-final-lora-Q8_0.gguf",
"lora_scale": 0.45,
"context_tokens": 131072,
"web_ui": "http://127.0.0.1:8091",
"prompt_embedded_in_template": true,
"prompt_sha256": "5591AC2C99FDC6C8E21D1D3B87625BB368204548BEC7B63C3FD6598CF23639BF",
"mode": "llama.cpp text Web UI; native start.ps1 retains text/image/audio/video"
}
}
|