Instructions to use King3Djbl/FableForge-1.5B-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use King3Djbl/FableForge-1.5B-v2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="King3Djbl/FableForge-1.5B-v2")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("King3Djbl/FableForge-1.5B-v2")
model = AutoModelForCausalLM.from_pretrained("King3Djbl/FableForge-1.5B-v2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use King3Djbl/FableForge-1.5B-v2 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 King3Djbl/FableForge-1.5B-v2:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M

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 King3Djbl/FableForge-1.5B-v2:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M

Use Docker

docker model run hf.co/King3Djbl/FableForge-1.5B-v2:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use King3Djbl/FableForge-1.5B-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "King3Djbl/FableForge-1.5B-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "King3Djbl/FableForge-1.5B-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/King3Djbl/FableForge-1.5B-v2:Q4_K_M

SGLang

How to use King3Djbl/FableForge-1.5B-v2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "King3Djbl/FableForge-1.5B-v2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "King3Djbl/FableForge-1.5B-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "King3Djbl/FableForge-1.5B-v2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "King3Djbl/FableForge-1.5B-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use King3Djbl/FableForge-1.5B-v2 with Ollama:
```
ollama run hf.co/King3Djbl/FableForge-1.5B-v2:Q4_K_M
```
Unsloth Desktop

How to use King3Djbl/FableForge-1.5B-v2 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M

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": "King3Djbl/FableForge-1.5B-v2:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use King3Djbl/FableForge-1.5B-v2 with Docker Model Runner:
```
docker model run hf.co/King3Djbl/FableForge-1.5B-v2:Q4_K_M
```

Lemonade

How to use King3Djbl/FableForge-1.5B-v2 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull King3Djbl/FableForge-1.5B-v2:Q4_K_M

Run and chat with the model

lemonade run user.FableForge-1.5B-v2-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use King3Djbl/FableForge-1.5B-v2 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M

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 King3Djbl/FableForge-1.5B-v2:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use King3Djbl/FableForge-1.5B-v2 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M

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 "King3Djbl/FableForge-1.5B-v2:Q4_K_M" \
  --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"

Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

FableForge-1.5B-v2

Continued training iteration — pushed further on the full NEXUS corpus.

· Fine-tuned from: FableForge-1.5B (v1) · Part of the FableForge ecosystem.

⚡ Quickest start — Ollama

ollama run hf.co/King3Djbl/FableForge-1.5B-v2:Q4_K_M

🦙 llama.cpp

llama-cli -hf King3Djbl/FableForge-1.5B-v2:Q4_K_M -p "Hello!"

💠 LM Studio

Search King3Djbl/FableForge-1.5B-v2 and pick a quant below.

📦 Provided quants (download one file, not the whole repo)

File	Quant	Size	Notes
FableForge-1.5B-v2-f16.gguf	`F16`	3.09 GB	Full precision (unquantized). Reference / conversion.
FableForge-1.5B-v2-q8_0.gguf	`Q8_0`	1.65 GB	Extremely high quality — usually overkill.
FableForge-1.5B-v2-q6_k.gguf	`Q6_K`	1.27 GB	Very high quality, near-perfect. Recommended if you have the RAM.
FableForge-1.5B-v2-q5_k_m.gguf	`Q5_K_M`	1.13 GB	High quality. Recommended.
FableForge-1.5B-v2-q4_k_m.gguf	`Q4_K_M`	0.99 GB	⭐ Best size/quality balance — default pick for most users.
FableForge-1.5B-v2-iq4_xs.gguf	`IQ4_XS`	0.90 GB	imatrix — great quality, smaller than Q4_K_S. Recommended for low RAM.
FableForge-1.5B-v2-q4_0.gguf	`Q4_0`	0.93 GB	Legacy format — prefer Q4_K_M.
FableForge-1.5B-v2-q3_k_m.gguf	`Q3_K_M`	0.82 GB	Lower quality; usable when RAM is tight.
FableForge-1.5B-v2-iq3_xxs.gguf	`IQ3_XXS`	0.67 GB	imatrix very low quality.
FableForge-1.5B-v2-q2_k.gguf	`Q2_K`	0.68 GB	Very low quality — surprisingly usable.
FableForge-1.5B-v2-iq2_xxs.gguf	`IQ2_XXS`	0.51 GB	Tiny imatrix — last resort.

Not sure? Take Q4_K_M. Low RAM → IQ4_XS. Max quality → Q6_K.

🧠 Prompt format (ChatML)

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

🧪 Performance

Non-NEXUS Benchmark — 45 prompts across 5 categories (general knowledge, uncensored, reasoning, tool use, hardware commands):

Category	v1 (1.5B)	v2 (1.5B)	ReasonCritic-7B
General Knowledge	47/50 (4.7)	49/50 (4.9)	47/50 (4.7)
Uncensored	48/50 (4.8)	33/50 (3.3)	41/50 (4.1)
Reasoning	50/50 (5.0)	44/50 (4.4)	49/50 (4.9)
Tool Use	49/50 (4.9)	38/50 (3.8)	49/50 (4.9)
Hardware Commands	23/25 (4.6)	22/25 (4.4)	25/25 (5.0)
Overall	217/225 (96.4%)	186/225 (82.7%)	211/225 (93.8%)

Benchmark: 45 curated prompts across 5 categories, scored 0-5 per response.

⭐ Like & share — it helps people find the source instead of a mirror.

Downloads last month: 139

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for King3Djbl/FableForge-1.5B-v2

Base model

fableforge-ai/ShellWhisperer-1.5B

Finetuned

King3Djbl/FableForge-1.5B

Finetuned

(1)

this model

Quantizations

1 model