Instructions to use nebada1101/Nexus-VMC-v1.4-RCL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nebada1101/Nexus-VMC-v1.4-RCL with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="nebada1101/Nexus-VMC-v1.4-RCL",
	filename="nexus-vmc-v1.4.Q8_0.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use nebada1101/Nexus-VMC-v1.4-RCL with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0
# Run inference directly in the terminal:
llama-cli -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0
# Run inference directly in the terminal:
llama-cli -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Use Docker

docker model run hf.co/nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

LM Studio
Jan
Ollama
How to use nebada1101/Nexus-VMC-v1.4-RCL with Ollama:
```
ollama run hf.co/nebada1101/Nexus-VMC-v1.4-RCL:Q8_0
```

Unsloth Studio new

How to use nebada1101/Nexus-VMC-v1.4-RCL with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for nebada1101/Nexus-VMC-v1.4-RCL to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for nebada1101/Nexus-VMC-v1.4-RCL to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for nebada1101/Nexus-VMC-v1.4-RCL to start chatting

Pi new

How to use nebada1101/Nexus-VMC-v1.4-RCL with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "nebada1101/Nexus-VMC-v1.4-RCL:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use nebada1101/Nexus-VMC-v1.4-RCL with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf nebada1101/Nexus-VMC-v1.4-RCL: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://127.0.0.1:8080/v1
hermes config set model.default nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Run Hermes

hermes

Docker Model Runner
How to use nebada1101/Nexus-VMC-v1.4-RCL with Docker Model Runner:
```
docker model run hf.co/nebada1101/Nexus-VMC-v1.4-RCL:Q8_0
```

Lemonade

How to use nebada1101/Nexus-VMC-v1.4-RCL with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull nebada1101/Nexus-VMC-v1.4-RCL:Q8_0

Run and chat with the model

lemonade run user.Nexus-VMC-v1.4-RCL-Q8_0

List all available models

lemonade list

nebada1101 commited on Apr 7

Commit

a46eb39

verified ·

1 Parent(s): bc76933

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +64 -3

README.md CHANGED Viewed

@@ -1,3 +1,64 @@
----
-license: apache-2.0
----

+---
+license: apache-2.0
+base_model: Qwen/Qwen2.5-1.5B
+tags:
+- medical
+- tool-use
+- reasoning
+- excel
+- assistant
+- vmc-rcl
+model_name: Nexus-VMC-v1.4-RCL
+---
+# Nexus-VMC-v1.4-RCL (1.5B) - Unified Strategic Intelligence
+Nexus-VMC-v1.4-RCL is a high-performance Small Language Model (SLM) meticulously fine-tuned for the **Antigravity Hub**. It excels in medical reasoning, Excel data logic, and complex tool-calling via the **VMC-RCL (Reasoning-Call-Logic)** protocol.
+## Key Features
+- **Strategic Reasoning**: Native support for `<vmc_think>` blocks, enabling deep analytical processing before task execution.
+- **Precision Tool-Calling**: 100% adherence to `[TOOL_CALL]` JSON schemas in comparative benchmarking.
+- **Unified Domain Mastery**: Expert-level performance in Medical clinical reasoning, Excel formula generation (SUMIF, etc.), and Document RAG.
+- **Ultra-Efficient Inference**: Optimized for local deployment on 8GB RAM environments (Laptop GPUs, CPU inference).
+## Benchmark Results (Phase 1.4 RCL)
+Benchmarked against SLM baselines and 9B-class models on a 4070 Laptop GPU (Ollama API).
+| Model | Avg TPS | Med Reasoning | Excel Logic | Hub Protocol Adherence |
+| :--- | :--- | :--- | :--- | :--- |
+| **Nexus-VMC-v1.4-RCL** | **130.3** | **PASS (Strategic)** | **PASS (SUMIF)** | **Native (S-Tier)** |
+| hauhau-qwen9b | 37.2 | PASS (Text) | PASS (Text) | B-Tier |
+| phi3.5:latest | 94.1 | PARTIAL (Python) | PASS (Formula) | C-Tier |
+| qwen2.5:1.5b | 196.4 | FAIL (Manual) | FAIL (Repeated) | D-Tier |
+## Usage Examples
+### VMC-RCL Protocol
+```text
+User: 병원의 재원 할당을 최적화하기 위해 현재 환자 수를 분석해줘.
+Assistant:
+<vmc_think>
+환자 수 데이터를 기반으로 한 최적화 분석이 필요함.
+hospital_warehouse 도구를 호출하여 실시간 데이터를 가져오는 논리적 흐름 설정.
+</vmc_think>
+[TOOL_CALL]
+{"name": "mcp_tool", "kwargs": {"target": "hospital_warehouse", "query": "current_patient_census"}}
+```
+## How to Run locally (Ollama)
+1. Download the `nexus-vmc-v1.4.Q8_0.gguf` file.
+2. Create a `Modelfile`:
+   ```text
+   FROM ./nexus-vmc-v1.4.Q8_0.gguf
+   TEMPLATE "<|im_start|>user\n{{ .Prompt }}<|im_end|>\n<|im_start|>assistant\n"
+   PARAMETER stop "<|im_end|>"
+   ```
+3. Run `ollama create nexus-v1.4 -f Modelfile`
+4. Run `ollama run nexus-v1.4`
+## License
+[Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)