Instructions to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-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 kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-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 kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
llama cli -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
llama cli -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

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 kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
./llama-cli -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

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 kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
./build/bin/llama-cli -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

Use Docker

docker model run hf.co/kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

LM Studio
Jan

vLLM

How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-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": "kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

Ollama
How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF with Ollama:
```
ollama run hf.co/kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP
```
Unsloth Desktop

How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

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": "kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF with Docker Model Runner:
```
docker model run hf.co/kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP
```

Lemonade

How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

Run and chat with the model

lemonade run user.Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF-Q4_0_ROCMFP

List all available models

lemonade list

Hermes Agent

How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-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 kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

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 kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP

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 "kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF:Q4_0_ROCMFP" \
  --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"

Granite 4.2-8B (STRIX_LEAN) — ROCmFP4 for AMD Strix Halo (gfx1151)

I built this STRIX_LEAN quantization of ibm-granite/granite-4.2-8b on my Strix Halo box for the ROCmFPX runtime. This is the 4th tier of my Granite 4.2 set — the lean 4-bit one people normally want.

The file


ftype	`106` — `Q4_0_ROCMFP4_STRIX_LEAN`
size	4,868,784,192 bytes (4.53 GiB)
bpw	4.43
architecture	`granite`
tensors	363
context	131,072
token embedding	Q5_K (the LEAN part)
`output.weight`	Q6_K (protected)
sha256	`0725133866578f418b82d0d3e7dc9ff43bc8387d89eb190ec642a9f04995c0a4`

Type histogram, read from the finished file:

Q4_0_ROCMFP4_FAST x200, F32 x81, Q4_0_ROCMFP4 x80, Q6_K x1, Q5_K x1

What STRIX_LEAN is — and what it protects

STRIX_LEAN is my lean 4-bit tier. The body is ROCmFP4 with the Strix Halo attention K/V quality recipe (that is what the STRIX part buys you), and the token embedding table is trimmed to Q5_K — that is the LEAN part, the size saving versus my COHERENT tier, which keeps the embeddings at Q6_K.

What never gets trimmed is the head. Every STRIX_LEAN I publish carries the protected Q6_K LM head. This model has tie_word_embeddings: false, so output.weight is a real standalone tensor, and a 4-bit head would degrade the logits of every single token. I quantized with --output-tensor-type q6_K and confirmed the head landed at Q6_K by exact-name read-back on the finished file (output.weight — exact match, not substring).

How I built it

Manifest gate: pulled ibm-granite/granite-4.2-8b file list from the HF API with ?blobs=true and recorded the real shard bytes (4 safetensors shards, 17,583,228,032 bytes total — never the index total_size).
Downloaded and byte-verified all 17 files against that manifest (sizes + LFS sha256).
Converted with convert_hf_to_gguf.py from my rocmfpx-dspark-halo tree (4eca07e), --outtype bf16 → 363 tensors, 17,587,421,248 bytes.
Quantized with the same tree's llama-quantize at 16 threads with --output-tensor-type q6_K. Dry-run estimate 4,639.83 MiB (4.43 bpw); the real file landed within ~3.5 MiB of it.

Measured on my box — full GPU offload

amd-halo: AMD Ryzen AI Max+ 395 (Strix Halo, gfx1151), ROCm 7.13.0, 128 GiB unified memory. Functional check at full offload — server flags -dev ROCm0 -fa on -ngl 999 --no-mmap -fit off -np 1 -b 2048 -c 8192 -t 16 --jinja, port 8497, greedy. 8 other llama-server seats were live on this machine while I tested (MemAvailable 16.1 GiB before load → 9.7 GiB after), so this is a functional check, not an idle-box benchmark.


offload	FULL — server log: `offloaded 41/41 layers to GPU`, GTT usage +5.87 GB on load
generation (server-reported)	38.33 t/s over 128 tokens
prompt processing	19 tokens in 108.2 ms

Sample output (greedy, prompt "Explain in one clear sentence what granite rock is primarily made of."):

Answer: Granite is primarily made of quartz, feldspar, and mica minerals. … (continued in the model's native self-check scaffold — real, structured generation)

⚠️ Stock llama.cpp will not load this file

Q4_0_ROCMFP4_STRIX_LEAN is a custom tensor format that exists only in the ROCmFPX fork of llama.cpp.

llama-server -m granite-4.2-8b-Q4_0_ROCMFP4_STRIX_LEAN.gguf -dev ROCm0 -fa on -ngl 999 -c 8192

Not measured

No benchmark sweeps, no context sweeps, no perplexity — one full-offload functional check, per my build discipline.

Provenance & license

Converted and quantized from ibm-granite/granite-4.2-8b (Apache 2.0). This quantized build is released under the same Apache 2.0 license. The ROCmFPX runtime is a third-party fork; its own terms apply to the runtime, not to these weights.

All my quants of Granite-4.2-8B

build	what it is	size	tok/s (full GPU offload)
`STRIX_LEAN`	my leaner 4-bit tier, Q6_K head — smallest of my 4-bit builds, the one most people want	4.53 GiB	38.33
`COHERENT`	my 4-bit ROCmFP4 tier with the Q6_K-protected head — the balance I run day to day	4.81 GiB	39.35
`Q8_0`	straight 8-bit ROCmFPX — highest fidelity I publish	8.45 GiB	21.49
`Q8_0-AGENT`	8-bit ROCmFPX with the agent-tuned tensor set — for tool-calling work where precision matters	8.60 GiB	23.38

All measured by me on a Ryzen AI MAX+ 395 (Strix Halo, gfx1151, ROCm 7.2.4) with the whole model on GPU (-ngl 999), 128-token greedy generation. A dash means I haven't measured that one yet — I won't put a number in a card I didn't measure.

Base model: ibm-granite/granite-4.2-8b

Downloads last month: 65

GGUF

Model size

9B params

Architecture

granite

Hardware compatibility

4-bit

Model tree for kingjones777/Granite-4.2-8B-ROCmFP4-STRIX_LEAN-GGUF

Base model

ibm-granite/granite-4.1-8b-base

Finetuned

ibm-granite/granite-4.2-8b

Quantized

(37)

this model