Text Generation
GGUF
English
Vietnamese
llama.cpp
on-device
mobile
ios
macos
qwen3.5
automatic-speech-recognition
asr
speech
qwen3-asr
conversational
Instructions to use haohan84/darksurf-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use haohan84/darksurf-models with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf haohan84/darksurf-models:UD-Q4_K_XL # Run inference directly in the terminal: llama cli -hf haohan84/darksurf-models:UD-Q4_K_XL
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf haohan84/darksurf-models:UD-Q4_K_XL # Run inference directly in the terminal: llama cli -hf haohan84/darksurf-models:UD-Q4_K_XL
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf haohan84/darksurf-models:UD-Q4_K_XL # Run inference directly in the terminal: ./build/bin/llama cli -hf haohan84/darksurf-models:UD-Q4_K_XL
Use Docker
docker model run hf.co/haohan84/darksurf-models:UD-Q4_K_XL
- Llama new
- LM Studio
- Jan
- vLLM
How to use haohan84/darksurf-models with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "haohan84/darksurf-models" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "haohan84/darksurf-models", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/haohan84/darksurf-models:UD-Q4_K_XL
- Ollama
How to use haohan84/darksurf-models with Ollama:
ollama run hf.co/haohan84/darksurf-models:UD-Q4_K_XL
- Unsloth Desktop
- Docker Model Runner
How to use haohan84/darksurf-models with Docker Model Runner:
docker model run hf.co/haohan84/darksurf-models:UD-Q4_K_XL
- Lemonade
How to use haohan84/darksurf-models with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull haohan84/darksurf-models:UD-Q4_K_XL
Run and chat with the model
lemonade run user.darksurf-models-UD-Q4_K_XL
List all available models
lemonade list
- Atomic Chat
Download catalog.json from haohan84/darksurf-models: direct link, hf CLI and curl.
- Browser
- Download file 2.65 kB
-
https://huggingface.co/haohan84/darksurf-models/resolve/main/catalog.json
- Command line
-
hf download hf://haohan84/darksurf-models/catalog.json
-
curl -L -o catalog.json https://huggingface.co/haohan84/darksurf-models/resolve/main/catalog.json
2.65 kB
| { | |
| "comment": "On-device models DarkSurf offers in Settings. Pinned to a Hugging Face commit; the app checks size and SHA-256 after download. Apache-2.0 (Qwen3.5), GGUF by Unsloth.", | |
| "version": 4, | |
| "feed": "https://huggingface.co/haohan84/darksurf-models/resolve/main/catalog.json", | |
| "models": [ | |
| { | |
| "id": "qwen3.5-0.8b-ud-q4kxl", | |
| "name": "Qwen3.5 0.8B", | |
| "quantization": "UD-Q4_K_XL", | |
| "file": "Qwen3.5-0.8B-UD-Q4_K_XL.gguf", | |
| "url": "https://huggingface.co/haohan84/darksurf-models/resolve/38fd158fb243406e6d19a5068d7345bde28c6735/Qwen3.5-0.8B-UD-Q4_K_XL.gguf", | |
| "size": 558772480, | |
| "sha256": "3177ebd67afe4438374da19e690bc1b98756f7e0fea9240e1be404336156a7b5", | |
| "minMemoryGB": 3.0, | |
| "note": "Fast on every device. The default." | |
| }, | |
| { | |
| "id": "qwen3.5-2b-ud-q4kxl", | |
| "name": "Qwen3.5 2B", | |
| "quantization": "UD-Q4_K_XL", | |
| "file": "Qwen3.5-2B-UD-Q4_K_XL.gguf", | |
| "url": "https://huggingface.co/haohan84/darksurf-models/resolve/9ff1e30b03aaa767e9c4e2afe0526b07b9ff5374/Qwen3.5-2B-UD-Q4_K_XL.gguf", | |
| "size": 1339752704, | |
| "sha256": "0af96165ea615bea39a04118d63f0b6d35908aea850ee4a51aa6151d851b8b35", | |
| "minMemoryGB": 6.0, | |
| "note": "Understands photos and reads text in them, answers better than the 0.8B. For phones with 6 GB of RAM or more.", | |
| "mmproj": { | |
| "file": "mmproj-F16.gguf", | |
| "url": "https://huggingface.co/haohan84/darksurf-models/resolve/a570ba3a26ce1b30a05715dcb1f1f1605c6550e2/mmproj-F16.gguf", | |
| "size": 668227264, | |
| "sha256": "7035e9cb8d7c6a9681d07eef9a364783e86ea4cd73faab2eabb4f43a101830c7" | |
| } | |
| } | |
| ], | |
| "speech": [ | |
| { | |
| "id": "qwen3-asr-0.6b-q8", | |
| "name": "Qwen3-ASR 0.6B", | |
| "quantization": "Q8_0", | |
| "minMemoryGB": 3.0, | |
| "note": "Speech to text in 52 languages, including Vietnamese. Runs faster than real time.", | |
| "files": [ | |
| { | |
| "role": "model", | |
| "file": "Qwen3-ASR-0.6B-Q8_0.gguf", | |
| "url": "https://huggingface.co/haohan84/darksurf-models/resolve/d4848b92bc44afb798b8176bf53c3c0d6bbe2d1b/Qwen3-ASR-0.6B-Q8_0.gguf", | |
| "size": 639442752, | |
| "sha256": "1708c32df43162e41e41ce69236530748b18a00465a0babc52b8bd21cd70f8dc" | |
| }, | |
| { | |
| "role": "mmproj", | |
| "file": "mmproj-Qwen3-ASR-0.6B-Q8_0.gguf", | |
| "url": "https://huggingface.co/haohan84/darksurf-models/resolve/a8419a61f909516f1bc478ec67193dbcf4590b4c/mmproj-Qwen3-ASR-0.6B-Q8_0.gguf", | |
| "size": 207941184, | |
| "sha256": "51ad77116cf33d90f46eb7c8b5ec30a3de2aa727e0888d2acb283d1465893b6e" | |
| } | |
| ] | |
| } | |
| ] | |
| } | |