Instructions to use gonibeed/Inkling-Small-grader-seed1234-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 gonibeed/Inkling-Small-grader-seed1234-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 gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf gonibeed/Inkling-Small-grader-seed1234-GGUF: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 gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M

Use Docker

docker model run hf.co/gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use gonibeed/Inkling-Small-grader-seed1234-GGUF with Ollama:
```
ollama run hf.co/gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M
```
Unsloth Desktop

How to use gonibeed/Inkling-Small-grader-seed1234-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gonibeed/Inkling-Small-grader-seed1234-GGUF: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": "gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use gonibeed/Inkling-Small-grader-seed1234-GGUF with Docker Model Runner:
```
docker model run hf.co/gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M
```

Lemonade

How to use gonibeed/Inkling-Small-grader-seed1234-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Inkling-Small-grader-seed1234-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use gonibeed/Inkling-Small-grader-seed1234-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 gonibeed/Inkling-Small-grader-seed1234-GGUF: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 gonibeed/Inkling-Small-grader-seed1234-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use gonibeed/Inkling-Small-grader-seed1234-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gonibeed/Inkling-Small-grader-seed1234-GGUF: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 "gonibeed/Inkling-Small-grader-seed1234-GGUF: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"

Access to this model is reviewed individually

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

These weights are fine-tuned on Python/SQL coursework. Access is granted case-by-case. Redistribution of the weights or any derivative is not permitted.

Inkling-Small grader (seed 1234) — Q4_K_M GGUF

Full merge of the private LoRA adapter gonibeed/inkling-small-grader-sft-v2 into thinkingmachines/Inkling-Small (bf16, 532 GB, ~265B MoE / ~12B active), then converted to GGUF and quantized to Q4_K_M.

Build provenance

base: thinkingmachines/Inkling-Small (apache-2.0)
adapter: r=32, alpha=32, 348 steps, lr 2e-4, seed 1234, val loss 10440 -> 7967
merge: tinker_train/merge_lora.py, 455 adapter modules -> 375 base tensors, scale alpha/r = 1.0, 673 tensors passed through
layout conventions: --w13-layout interleaved, --shared-layout expert-major (see merge_manifest.json)
llama.cpp: danielhanchen/llama.cpp branch add-inkling @ 49c0377bb07e
quantization path: bf16 -> GGUF q8_0 -> Q4_K_M (--allow-requantize)

Usage contract — this model was trained a specific way

Empty system prompt. The grading instruction lives in the user message.
Response contract: VERDICT: / FEEDBACK: (see eval/config.py::RESPONSE_CONTRACT)
reasoning_effort 0.9 — the renderer baked Thinking effort level: 0.9 into every training example, so 0.9 is the training distribution.

Trained on Python/SQL coursework. Access is granted on request.

Downloads last month: -

GGUF

Model size

264B params

Architecture

inkling

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for gonibeed/Inkling-Small-grader-seed1234-GGUF

Base model

thinkingmachines/Inkling-Small

Quantized

(54)

this model