Instructions to use oxbyte8/SparkBotLLM-1.7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oxbyte8/SparkBotLLM-1.7B with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("LiquidAI/LFM3-1.7B")
model = PeftModel.from_pretrained(base_model, "oxbyte8/SparkBotLLM-1.7B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use oxbyte8/SparkBotLLM-1.7B 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 oxbyte8/SparkBotLLM-1.7B:Q4_0
# Run inference directly in the terminal:
llama cli -hf oxbyte8/SparkBotLLM-1.7B:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf oxbyte8/SparkBotLLM-1.7B:Q4_0
# Run inference directly in the terminal:
llama cli -hf oxbyte8/SparkBotLLM-1.7B:Q4_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 oxbyte8/SparkBotLLM-1.7B:Q4_0
# Run inference directly in the terminal:
./build/bin/llama cli -hf oxbyte8/SparkBotLLM-1.7B:Q4_0

Use Docker

docker model run hf.co/oxbyte8/SparkBotLLM-1.7B:Q4_0

Llama new
LM Studio
Jan

vLLM

How to use oxbyte8/SparkBotLLM-1.7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "oxbyte8/SparkBotLLM-1.7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oxbyte8/SparkBotLLM-1.7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/oxbyte8/SparkBotLLM-1.7B:Q4_0

Ollama
How to use oxbyte8/SparkBotLLM-1.7B with Ollama:
```
ollama run hf.co/oxbyte8/SparkBotLLM-1.7B:Q4_0
```
Unsloth Desktop

How to use oxbyte8/SparkBotLLM-1.7B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf oxbyte8/SparkBotLLM-1.7B:Q4_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:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "oxbyte8/SparkBotLLM-1.7B:Q4_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use oxbyte8/SparkBotLLM-1.7B with Docker Model Runner:
```
docker model run hf.co/oxbyte8/SparkBotLLM-1.7B:Q4_0
```

Lemonade

How to use oxbyte8/SparkBotLLM-1.7B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull oxbyte8/SparkBotLLM-1.7B:Q4_0

Run and chat with the model

lemonade run user.SparkBotLLM-1.7B-Q4_0

List all available models

lemonade list

Hermes Agent

How to use oxbyte8/SparkBotLLM-1.7B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf oxbyte8/SparkBotLLM-1.7B:Q4_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 oxbyte8/SparkBotLLM-1.7B:Q4_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use oxbyte8/SparkBotLLM-1.7B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf oxbyte8/SparkBotLLM-1.7B:Q4_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://127.0.0.1:8080/v1 \
  --custom-model-id "oxbyte8/SparkBotLLM-1.7B:Q4_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"

SparkBotLLM-1.7B / training-report.json

oxbyte8

Publish initial SparkBotLLM Q4 preview and adaptation report

51ad3e9 verified 3 days ago

Raw History Blame Contribute Delete

3.15 kB

{
  "baseModel": "LiquidAI/LFM3-1.7B",
  "baseRevision": "ac2586a34f1a1c03329dbdcdefabd8b6478ddaa7",
  "stepsCompleted": 192,
  "requestedSteps": 192,
  "selectedCheckpointStep": 128,
  "selection": "Lowest domain validation loss; validation was used for checkpoint selection, not an independent final test.",
  "trainingExamples": 242,
  "heldoutDomainExamples": 6,
  "seed": 20261030,
  "loraRank": 8,
  "learningRate": 5e-05,
  "baseline": {
    "heldoutLoss": 4.622275193532308,
    "tools": [
      {
        "id": "read_gpu_clocks-heldout",
        "expectedTool": "read_gpu_clocks",
        "passed": true,
        "response": "<tool_call>\n<function=read_gpu_clocks>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "inspect_sparks-heldout",
        "expectedTool": "inspect_sparks",
        "passed": true,
        "response": "<tool_call>\n<function=inspect_sparks>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "check_setup-heldout",
        "expectedTool": "check_setup",
        "passed": true,
        "response": "<tool_call>\n<function=check_setup>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "list_recipes-heldout",
        "expectedTool": "list_recipes",
        "passed": true,
        "response": "<tool_call>\n<function=list_recipes>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "latest_models-heldout",
        "expectedTool": "latest_models",
        "passed": true,
        "response": "<tool_call>\n<function=latest_models>\n<parameter=task>\ntext-generation\n</parameter>\n</function>\n</tool_call><|im_end|>"
      }
    ]
  },
  "candidate": {
    "heldoutLoss": 3.3232267697652182,
    "tools": [
      {
        "id": "read_gpu_clocks-heldout",
        "expectedTool": "read_gpu_clocks",
        "passed": true,
        "response": "<tool_call>\n<function=read_gpu_clocks>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "inspect_sparks-heldout",
        "expectedTool": "inspect_sparks",
        "passed": true,
        "response": "<tool_call>\n<function=inspect_sparks>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "check_setup-heldout",
        "expectedTool": "check_setup",
        "passed": true,
        "response": "<tool_call>\n<function=check_setup>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "list_recipes-heldout",
        "expectedTool": "list_recipes",
        "passed": true,
        "response": "<tool_call>\n<function=list_recipes>\n</function>\n</tool_call><|im_end|>"
      },
      {
        "id": "latest_models-heldout",
        "expectedTool": "latest_models",
        "passed": true,
        "response": "<tool_call>\n<function=latest_models>\n<parameter=task>\ntext-generation\n</parameter>\n</function>\n</tool_call><|im_end|>"
      }
    ]
  },
  "stoppedReason": null,
  "seconds": 132.38,
  "cudaPeakGiB": 3.592775344848633,
  "productionHelperChanged": false,
  "published": false,
  "limits": "Small synthetic pilot. Held-out loss and five routing checks do not establish broad factual accuracy or production quality."
}

	{
	"baseModel": "LiquidAI/LFM3-1.7B",
	"baseRevision": "ac2586a34f1a1c03329dbdcdefabd8b6478ddaa7",
	"stepsCompleted": 192,
	"requestedSteps": 192,
	"selectedCheckpointStep": 128,
	"selection": "Lowest domain validation loss; validation was used for checkpoint selection, not an independent final test.",
	"trainingExamples": 242,
	"heldoutDomainExamples": 6,
	"seed": 20261030,
	"loraRank": 8,
	"learningRate": 5e-05,
	"baseline": {
	"heldoutLoss": 4.622275193532308,
	"tools": [
	{
	"id": "read_gpu_clocks-heldout",
	"expectedTool": "read_gpu_clocks",
	"passed": true,
	"response": "<tool_call>\n<function=read_gpu_clocks>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "inspect_sparks-heldout",
	"expectedTool": "inspect_sparks",
	"passed": true,
	"response": "<tool_call>\n<function=inspect_sparks>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "check_setup-heldout",
	"expectedTool": "check_setup",
	"passed": true,
	"response": "<tool_call>\n<function=check_setup>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "list_recipes-heldout",
	"expectedTool": "list_recipes",
	"passed": true,
	"response": "<tool_call>\n<function=list_recipes>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "latest_models-heldout",
	"expectedTool": "latest_models",
	"passed": true,
	"response": "<tool_call>\n<function=latest_models>\n<parameter=task>\ntext-generation\n</parameter>\n</function>\n</tool_call><\|im_end\|>"
	}
	]
	},
	"candidate": {
	"heldoutLoss": 3.3232267697652182,
	"tools": [
	{
	"id": "read_gpu_clocks-heldout",
	"expectedTool": "read_gpu_clocks",
	"passed": true,
	"response": "<tool_call>\n<function=read_gpu_clocks>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "inspect_sparks-heldout",
	"expectedTool": "inspect_sparks",
	"passed": true,
	"response": "<tool_call>\n<function=inspect_sparks>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "check_setup-heldout",
	"expectedTool": "check_setup",
	"passed": true,
	"response": "<tool_call>\n<function=check_setup>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "list_recipes-heldout",
	"expectedTool": "list_recipes",
	"passed": true,
	"response": "<tool_call>\n<function=list_recipes>\n</function>\n</tool_call><\|im_end\|>"
	},
	{
	"id": "latest_models-heldout",
	"expectedTool": "latest_models",
	"passed": true,
	"response": "<tool_call>\n<function=latest_models>\n<parameter=task>\ntext-generation\n</parameter>\n</function>\n</tool_call><\|im_end\|>"
	}
	]
	},
	"stoppedReason": null,
	"seconds": 132.38,
	"cudaPeakGiB": 3.592775344848633,
	"productionHelperChanged": false,
	"published": false,
	"limits": "Small synthetic pilot. Held-out loss and five routing checks do not establish broad factual accuracy or production quality."
	}