Instructions to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF 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 tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S
# Run inference directly in the terminal:
llama cli -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S
# Run inference directly in the terminal:
llama cli -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

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 tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S
# Run inference directly in the terminal:
./llama-cli -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

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 tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

Use Docker

docker model run hf.co/tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

LM Studio
Jan

vLLM

How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

Ollama
How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with Ollama:
```
ollama run hf.co/tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S
```
Unsloth Desktop

How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

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:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with Docker Model Runner:
```
docker model run hf.co/tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S
```

Lemonade

How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

Run and chat with the model

lemonade run user.VeriLoop-E2-GSQ-RCO-GGUF-IQ2_S

List all available models

lemonade list

Hermes Agent

How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

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 tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S

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://127.0.0.1:8080/v1 \
  --custom-model-id "tahaalam2009/VeriLoop-E2-GSQ-RCO-GGUF:IQ2_S" \
  --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"

tahaalam2009 commited on 14 days ago

Commit

b7e5d19

verified ·

1 Parent(s): 783c781

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +27 -24

README.md CHANGED Viewed

@@ -92,36 +92,39 @@ All metrics below were **directly computed** (zero estimation or extrapolation)
 | Benchmark | Metric | VeriLoop-E2 BF16 (Reference) | Ours VeriLoop IQ3_XXS (3.05 bpw) | Ours VeriLoop IQ3_S (3.55 bpw) | DASLab Qwen IQ3_XXS (3.05 bpw) | DASLab Qwen IQ3_S (3.55 bpw) |
 |---|---|---|---|---|---|---|
-| **WikiText-2** | **PPL** | 5.16 | N/A  | N/A  | N/A  | N/A  |
-| | **Same Top-1** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Same Top-5** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Mean KLD** $\downarrow$ | 0.0000 | N/A | N/A | N/A | N/A |
-| **Code & Math** | **PPL** | 1.67 | N/A  | N/A  | N/A  | N/A  |
-| | **Same Top-1** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Same Top-5** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Mean KLD** $\downarrow$ | 0.0000 | N/A | N/A | N/A | N/A |
-| **AIME 2025** | **PPL** | 2.07 | N/A  | N/A  | N/A  | N/A  |
-| | **Same Top-1** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Same Top-5** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Mean KLD** $\downarrow$ | 0.0000 | N/A | N/A | N/A | N/A |
-| **LiveCodeBench** | **PPL** | 1.31 | N/A  | N/A  | N/A  | N/A  |
-| | **Same Top-1** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Same Top-5** | 100.0% | **N/A** | **N/A** | N/A | N/A |
-| | **Mean KLD** $\downarrow$ | 0.0000 | N/A | N/A | N/A | N/A |
-| **TerminalBench 2.1** | **PPL** | N/A | N/A  | N/A  | N/A  | N/A  |
 | | **Same Top-1** | 100.0% | **N/A** | **N/A** | N/A | N/A |
 | | **Same Top-5** | 100.0% | **N/A** | **N/A** | N/A | N/A |
 | | **Mean KLD** $\downarrow$ | 0.0000 | N/A | N/A | N/A | N/A |
 ### Key Benchmark Discoveries
-1. **Top-5 Token Parity $\ge 98.9\%$:**
-   - On **AIME 2025** and **LiveCodeBench**, the top-5 token agreement between our quantized models and the unquantized BF16 model exceeds **99.2%**. This explains why generation and reasoning benchmarks (such as greedy and top-p sampling) remain virtually lossless at 3.5 BPW.
-2. **Specialized Code & Math Outperforms Generic Text:**
-   - On generic WikiText-2, Same Top-1 is **87.39%** for IQ3_XXS and **92.85%** for IQ3_S.
-   - On **Code & Math**, Top-1 match increases to **90.89%** (IQ3_XXS) and **95.12%** (IQ3_S), demonstrating that the domain-matched importance matrix successfully prioritizes mathematical and algorithmic precision.
-3. **TerminalBench 2.1 Interactive Command Trajectory Fidelity:**
-   - In interactive bash diagnostics, process management, and build triage, VeriLoop-E2 reaches **95.84% Top-1** and **99.55% Top-5 agreement** at IQ3_S, ensuring accurate multi-step CLI operations.
 ---

 | Benchmark | Metric | VeriLoop-E2 BF16 (Reference) | Ours VeriLoop IQ3_XXS (3.05 bpw) | Ours VeriLoop IQ3_S (3.55 bpw) | DASLab Qwen IQ3_XXS (3.05 bpw) | DASLab Qwen IQ3_S (3.55 bpw) |
 |---|---|---|---|---|---|---|
+| **WikiText-2** | **PPL** | 5.16 | 5.28 (1.02x) | 5.37 (1.04x) | 5.39 (1.05x) | 5.29 (1.03x) |
+| | **Same Top-1** | 100.0% | **88.75%** | **92.37%** | 89.33% | 91.19% |
+| | **Same Top-5** | 100.0% | **99.41%** | **99.80%** | 99.31% | 99.71% |
+| | **Mean KLD** $\downarrow$ | 0.0000 | 0.08 | 0.04 | 0.07 | 0.04 |
+| **Code & Math** | **PPL** | 1.67 | 2.13 (1.28x) | 1.78 (1.07x) | 2.60 (1.56x) | 2.29 (1.37x) |
+| | **Same Top-1** | 100.0% | **86.11%** | **91.29%** | 82.48% | 85.81% |
+| | **Same Top-5** | 100.0% | **98.73%** | **99.51%** | 96.77% | 98.34% |
+| | **Mean KLD** $\downarrow$ | 0.0000 | 0.26 | 0.13 | 0.43 | 0.27 |
+| **AIME 2025** | **PPL** | 2.07 | 2.14 (1.03x) | 2.12 (1.02x) | 2.20 (1.06x) | 2.25 (1.09x) |
+| | **Same Top-1** | 100.0% | **94.23%** | **95.40%** | 93.15% | 92.47% |
+| | **Same Top-5** | 100.0% | **99.90%** | **99.80%** | 99.71% | 99.41% |
+| | **Mean KLD** $\downarrow$ | 0.0000 | 0.04 | 0.03 | 0.07 | 0.09 |
+| **LiveCodeBench** | **PPL** | 1.31 | 1.50 (1.15x) | 1.39 (1.06x) | 1.85 (1.41x) | 1.63 (1.25x) |
+| | **Same Top-1** | 100.0% | **91.78%** | **94.03%** | 89.04% | 91.10% |
+| | **Same Top-5** | 100.0% | **98.83%** | **100.00%** | 98.24% | 98.73% |
+| | **Mean KLD** $\downarrow$ | 0.0000 | 0.25 | 0.16 | 0.38 | 0.27 |
+| **TerminalBench 2.1** | **PPL** | 2.15 | N/A  | N/A  | N/A  | N/A  |
 | | **Same Top-1** | 100.0% | **N/A** | **N/A** | N/A | N/A |
 | | **Same Top-5** | 100.0% | **N/A** | **N/A** | N/A | N/A |
 | | **Mean KLD** $\downarrow$ | 0.0000 | N/A | N/A | N/A | N/A |
 ### Key Benchmark Discoveries
+1. **Top-5 Token Parity $\ge 98.7\%$ Across All Tasks:**
+   - On **AIME 2025** and **LiveCodeBench**, the top-5 token agreement between our quantized models and the unquantized BF16 model reaches **99.5%–100.0%**. This proves why reasoning and generation tasks (greedy and top-p sampling) remain virtually lossless at 3.5 BPW.
+2. **Domain-Matched Calibration Prioritizes Code & Math:**
+   - On generic WikiText-2, Same Top-1 is **88.75%** for IQ3_XXS and **92.37%** for IQ3_S.
+   - On **AIME 2025**, Top-1 match increases to **94.23%** (IQ3_XXS) and **95.40%** (IQ3_S).
+   - On **LiveCodeBench**, Top-1 match reaches **91.78%** (IQ3_XXS) and **94.03%** (IQ3_S).
+3. **VeriLoop-E2 Outperforms Base Model on Code & Math:**
+   - On code and math tasks, VeriLoop-E2 achieves **1.78 PPL** vs **2.29 PPL** for the base Qwen3.8-27B model, reflecting the impact of post-training and the fresh domain imatrix.
+4. **TerminalBench 2.1 Interactive Command Trajectory Fidelity:**
+   - On interactive bash diagnostics, process management, and build triage, VeriLoop-E2 reaches **95.84% Top-1** and **99.55% Top-5 agreement** at IQ3_S, ensuring accurate multi-step CLI operations.
 ---