Instructions to use Harsha001/mammoth-1-ROCmFPX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Harsha001/mammoth-1-ROCmFPX 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 Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP
# Run inference directly in the terminal:
llama cli -hf Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP
# Run inference directly in the terminal:
llama cli -hf Harsha001/mammoth-1-ROCmFPX: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 Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP
# Run inference directly in the terminal:
./llama-cli -hf Harsha001/mammoth-1-ROCmFPX: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 Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP

Use Docker

docker model run hf.co/Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP

Llama new
LM Studio
Jan
Ollama
How to use Harsha001/mammoth-1-ROCmFPX with Ollama:
```
ollama run hf.co/Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP
```
Unsloth Desktop

How to use Harsha001/mammoth-1-ROCmFPX with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Harsha001/mammoth-1-ROCmFPX: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": "Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Harsha001/mammoth-1-ROCmFPX with Docker Model Runner:
```
docker model run hf.co/Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP
```

Lemonade

How to use Harsha001/mammoth-1-ROCmFPX with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP

Run and chat with the model

lemonade run user.mammoth-1-ROCmFPX-Q4_0_ROCMFP

List all available models

lemonade list

Hermes Agent

How to use Harsha001/mammoth-1-ROCmFPX with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Harsha001/mammoth-1-ROCmFPX: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 Harsha001/mammoth-1-ROCmFPX:Q4_0_ROCMFP

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Harsha001/mammoth-1-ROCmFPX with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Harsha001/mammoth-1-ROCmFPX: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 "Harsha001/mammoth-1-ROCmFPX: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"

Harsha001 commited on 2 days ago

Commit

3d9fbc4

verified ·

1 Parent(s): b4822f0

Rename the 6-bit file to Q6_K_ROCMFPX so the Hub lists it as 6-bit

Browse files

Files changed (3) hide show

.gitattributes +1 -0
README.md +3 -1
mammoth-1-Q6_0_ROCMFPX.gguf → mammoth-1-Q6_K_ROCMFPX.gguf +0 -0

.gitattributes CHANGED Viewed

@@ -37,3 +37,4 @@ mammoth-1-Q8_0_ROCMFPX.gguf filter=lfs diff=lfs merge=lfs -text
 mammoth-1-Q6_0_ROCMFPX.gguf filter=lfs diff=lfs merge=lfs -text
 mammoth-1-Q4_0_ROCMFP4_STRIX.gguf filter=lfs diff=lfs merge=lfs -text
 mammoth-1-Q4_0_ROCMFP4_FAST.gguf filter=lfs diff=lfs merge=lfs -text

 mammoth-1-Q6_0_ROCMFPX.gguf filter=lfs diff=lfs merge=lfs -text
 mammoth-1-Q4_0_ROCMFP4_STRIX.gguf filter=lfs diff=lfs merge=lfs -text
 mammoth-1-Q4_0_ROCMFP4_FAST.gguf filter=lfs diff=lfs merge=lfs -text
+mammoth-1-Q6_K_ROCMFPX.gguf filter=lfs diff=lfs merge=lfs -text

README.md CHANGED Viewed

@@ -33,10 +33,12 @@ text; they show how much each file changes the weights, not decision accuracy, w
 | File | Size | PPL | KLD | Same top | |
 |---|---:|---:|---:|---:|---|
 | `mammoth-1-Q8_0_ROCMFPX.gguf` | 27.8 GB | 6.8803 | 0.0058 | 97.26% | 8-bit, closest to BF16 |
-| `mammoth-1-Q6_0_ROCMFPX.gguf` | 22.2 GB | 6.8779 | 0.0065 | 96.74% | 6-bit, almost as close as 8-bit |
 | `mammoth-1-Q4_0_ROCMFP4_STRIX.gguf` | 14.8 GB | 6.9821 | 0.0370 | 90.98% | **Recommended.** 4-bit recipe tuned for Strix Halo (`gfx1151`) |
 | `mammoth-1-Q4_0_ROCMFP4_FAST.gguf` | 14.3 GB | 6.9557 | 0.0379 | 91.21% | smallest, speed-first 4-bit (one scale per block) |
 Not published: `Q3_0_ROCMFPX` (14.9 GB, PPL 7.0924, KLD 0.0549, same top 89.54%) is larger than both 4-bit files and
 further from BF16.

 | File | Size | PPL | KLD | Same top | |
 |---|---:|---:|---:|---:|---|
 | `mammoth-1-Q8_0_ROCMFPX.gguf` | 27.8 GB | 6.8803 | 0.0058 | 97.26% | 8-bit, closest to BF16 |
+| `mammoth-1-Q6_K_ROCMFPX.gguf` | 22.2 GB | 6.8779 | 0.0065 | 96.74% | 6-bit, almost as close as 8-bit |
 | `mammoth-1-Q4_0_ROCMFP4_STRIX.gguf` | 14.8 GB | 6.9821 | 0.0370 | 90.98% | **Recommended.** 4-bit recipe tuned for Strix Halo (`gfx1151`) |
 | `mammoth-1-Q4_0_ROCMFP4_FAST.gguf` | 14.3 GB | 6.9557 | 0.0379 | 91.21% | smallest, speed-first 4-bit (one scale per block) |
+The 6-bit file is named `mammoth-1-Q6_K_ROCMFPX.gguf` so that Hugging Face lists it as a 6-bit quant. Its weights are the ROCmFPX `Q6_0_ROCMFPX` type, not a stock llama.cpp `Q6_K`, and it needs the same fork as the other files.
 Not published: `Q3_0_ROCMFPX` (14.9 GB, PPL 7.0924, KLD 0.0549, same top 89.54%) is larger than both 4-bit files and
 further from BF16.

mammoth-1-Q6_0_ROCMFPX.gguf → mammoth-1-Q6_K_ROCMFPX.gguf RENAMED Viewed

File without changes