Instructions to use MigoXV/AgenticASR-Refiner-Q8_0 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 MigoXV/AgenticASR-Refiner-Q8_0 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 MigoXV/AgenticASR-Refiner-Q8_0:Q8_0 # Run inference directly in the terminal: llama cli -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0 # Run inference directly in the terminal: llama cli -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
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 MigoXV/AgenticASR-Refiner-Q8_0:Q8_0 # Run inference directly in the terminal: ./build/bin/llama cli -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Use Docker
docker model run hf.co/MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
- Llama new
- LM Studio
- Jan
- vLLM
How to use MigoXV/AgenticASR-Refiner-Q8_0 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MigoXV/AgenticASR-Refiner-Q8_0" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MigoXV/AgenticASR-Refiner-Q8_0", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
- Ollama
How to use MigoXV/AgenticASR-Refiner-Q8_0 with Ollama:
ollama run hf.co/MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
- Unsloth Desktop
- Pi
How to use MigoXV/AgenticASR-Refiner-Q8_0 with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "MigoXV/AgenticASR-Refiner-Q8_0:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use MigoXV/AgenticASR-Refiner-Q8_0 with Docker Model Runner:
docker model run hf.co/MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
- Lemonade
How to use MigoXV/AgenticASR-Refiner-Q8_0 with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Run and chat with the model
lemonade run user.AgenticASR-Refiner-Q8_0-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use MigoXV/AgenticASR-Refiner-Q8_0 with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:9931/v1 hermes config set model.default MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use MigoXV/AgenticASR-Refiner-Q8_0 with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf MigoXV/AgenticASR-Refiner-Q8_0:Q8_0
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "MigoXV/AgenticASR-Refiner-Q8_0:Q8_0" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
AgenticASR-Refiner Q8_0
ๅๅงๆจกๅ Andrew0425/AgenticASR-Refiner ็ Q8_0 GGUF ้ๅ็ๆฌ๏ผ็จไบ ASR ่ฝฌๅๆๆฌ็บ ้๏ผไธ็ดๆฅๅค็้ณ้ขใๆช่ฟฝๅ ่ฎญ็ปใ
ๆไปถไธ้ๅ
- ๆไปถ๏ผ
AgenticASR-Refiner-Q8_0.gguf๏ผ1,153,529,088 ๅญ่๏ผ็บฆ 1.07 GiB๏ผใ - SHA-256๏ผ
0cb29dcf5b84a45b0733cd50ac4ad1b879a4aafaa01f6d190c17fc3d9a10961fใ - ไปๅๅง FP32 checkpoint ๅฏผๅบ็ F16 GGUF ็ดๆฅ้ๅๅฐ Q8_0๏ผๆชๅฏน INT4 ๆ้ไบๆฌก้ๅใ
- 170 ไธช Q8_0 ็ฉ้ตใ49 ไธช F32 ๅผ ้๏ผ่พๅบๅคดไธบ Q8_0ใๅ็ผฉๆพๅ ๆฐๆฎไฝฟๆ้ๅนณๅ็บฆ 8.5 bit/ๅๆฐใ
- llama.cpp ๅบ็ก็ๆฌ๏ผ
f849095365d5ac1f560886a4e6574531d559e96fใๆฌๆบ CANN ่ฟ่กๆถๅฆๅซๅ ผๅฎน่กฅไธ๏ผ่ฟไธๆฏๅ็ OM ๆไปถใ head_dim=128๏ผrope_theta=5000000๏ผ่ฝฌๆขๆฅๆบๅ้ ็ฝฎ่งquantization.jsonใ
ไฝฟ็จ
ไฝฟ็จๆฏๆ่ฏฅๆจกๅ Llama ๆถๆๅ GGUF ็ llama.cppใๅ่ทฏ็คบไพ๏ผ
llama-server --model AgenticASR-Refiner-Q8_0.gguf \
--alias AgenticASR-Refiner --ctx-size 512 --parallel 1 \
--jinja --host 127.0.0.1 --port 8080
่พๅ ฅไธ่พๅบๅ ฑไบซไธไธๆใ512 token ๆฏๆฌๆฌก็ญๆๆฌๅทฅไฝ้ ็ฝฎ๏ผๅนถ้ๆจกๅ่ฎญ็ปไธไธๆไธ้ใๅปบ่ฎฎ็บฆ 128 ๅญ็่พๅ ฅใๆๅค 256 token ่พๅบ๏ผๅญๆฐไธ token ๆฐไธๅใ
่ฐ็จ Chat Completions ๆถๆพๅผไผ ๅ
ฅไปฅไธ system ๆ็คบ่ฏใๅๆก user ๆๆฌ๏ผ่ฎพ็ฝฎ temperature: 0ใmax_tokens: 256 ๅ stop: ["<|im_end|>"]๏ผ
ไฝ ๆฏ ASR ๆๆฌ็บ ้ๅฉๆใไฟ็ๅๆ๏ผๆๅฐไฟฎๆน๏ผๅปๅฃ็/้ๅค๏ผไฟฎ้ๅญ๏ผ่กฅๅฟ
่ฆๆ ็น๏ผ่ง่ๆฐๅญใๆฅๆใๆฏ่ฏญๅไปฃ็ ็ฌฆๅท๏ผๅค็่ชๆไฟฎๆญฃใไธ่ฆๆป็ปใๆฉๅๆ่งฃ้ใ้่ฆๆ้ๅฎไฝๅจๆซๅฐพ่ฟฝๅ <KEY>[่ฏ1ใ่ฏ2]๏ผๆฒกๆๅไธๅ ใ
ไธ่ฟฐ้็จๅฝไปคไฝฟ็จๅ
ๅต่ๅคฉๆจกๆฟ๏ผ2026-10-28 ็ๆฌๆบ้ช่ฏไฝฟ็จ็ญไปท็ๅ่ฝฎ็บ ้ๆจกๆฟ๏ผassistant ๅ็ผไธๆๅ
ฅๆ่ๆ ็ญพ๏ผๅนถ้ๆกๆฃๆฅ prompt token IDs ไธ FP32 ๅบ็บฟไธ่ดใ็งปๆคๅฐๅ
ถไป็ปๅฎๆ่ๅคฉๆจกๆฟๆถๅบๆฃๆฅ็ๆๅ็ผใ่พๅบไธญ็ <KEY>[...] ไธบๆจกๅ็ๆ็ๅฎไฝๅญๆฎตใ
ๅทฒๅฎๆ็ๆฌๆบ้ช่ฏ
ๆ่ พ 910B4-1ใCANN 9.1.1๏ผๅไธ llama.cpp CANN ๆๅปบ๏ผ่ดชๅฟ่งฃ็ ๏ผ
- 64 ๆก็ญๆๆฌๅ จ้จๆญฃๅธธๅๆญข๏ผ64/64 ่พๅบๆๆฌไธ Transformers CPU FP32 ๅบ็บฟ้ๅญ็ธๅ๏ผๅ็ป Q4_0 ไธบ 60/64ใ
- ๅไปฝๆ้ใ8 ไธชๆดปๅจๆงฝไฝ๏ผๆฏๆงฝ 512 token๏ผF16 KV ๅ ฑ 96 MiBใ
- 8 ่ทฏๅ 128 ๅญ่พๅ ฅๅ จ้จๆญฃๅธธ็ปๆ๏ผ8/8 ไธไธฒ่ก็ปๆ็ธๅ๏ผSSE ๆผๆฅไธๆฎ้ๅๅบไธ่ดใ
- NPU ๆจกๅ่ฟ็จๅฐฑ็ปช็บฆ 1213 MiB๏ผๆจ็้ๆ ทๆ้ซ 1267 MiBใๅๆๆๅ ถไป NPU ๆๅก๏ผๆชๅ้็ฆปๆง่ฝๅบๅๆ้ฟๆๅๅๆต่ฏใ
Q8 ไปๆฏๆๆ้ๅใๅฐๆ ทๆฌ่พๅบ็ธๅไธ็ญไบๅ
จ้ขๆ ๆ๏ผไนไธๆฏ็บ ้ๅ็กฎ็ใๅๅงๆจกๅ็้่ฏฏไปๅฏ่ฝไฟ็๏ผไพๅฆๆฌ็ปโๅไบไธไธๅๅ
โๅจ FP32 ไธ Q8 ไธญ้ฝ่ขซ้่ฏฏๅๆ 12,300ๅ
ใๅฎ้
้จ็ฝฒๅบไฝฟ็จไธๅกๆๆฌ่ฏไผฐๆฐๅญใๅฐๅๅไธๆๅ่ฏใ
ๆฅๆบไธ่ฎธๅฏ
ๅๅงๆจกๅ๏ผhttps://huggingface.co/Andrew0425/AgenticASR-Refiner
้ๅๆฒฟ็จไธๆธธ Apache-2.0 ่ฎธๅฏ๏ผ่ง LICENSEใ
- Downloads last month
- 38
8-bit
Model tree for MigoXV/AgenticASR-Refiner-Q8_0
Base model
Andrew0425/AgenticASR-Refiner