Instructions to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
llama cli -hf raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
llama cli -hf raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
./llama-cli -hf raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP
# Run inference directly in the terminal:
./build/bin/llama-cli -hf raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP

Use Docker

docker model run hf.co/raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP

LM Studio
Jan

vLLM

How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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": "raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP

Ollama
How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with Ollama:
```
ollama run hf.co/raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP
```

Unsloth Studio

How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF to start chatting

How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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 "raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-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"

Docker Model Runner
How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with Docker Model Runner:
```
docker model run hf.co/raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP
```

Lemonade

How to use raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull raulvidis/KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF:Q4_0_ROCMFP

Run and chat with the model

lemonade run user.KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF-Q4_0_ROCMFP

List all available models

lemonade list

KAT-Coder-V2.5-Dev-ROCmFP4-STRIX-MTP-GGUF / README.md

raulvidis

docs: update filename after adding Q4_0 quant token

61c993a verified 1 day ago

preview code

Raw

History Blame Contribute Delete

3.85 kB

	---
	license: apache-2.0
	base_model: Kwaipilot/KAT-Coder-V2.5-Dev
	base_model_relation: quantized
	tags:
	- gguf
	- rocm
	- rocmfp4
	- strix-halo
	- gfx1151
	- speculative-decoding
	- mtp
	pipeline_tag: text-generation
	---

	# KAT-Coder-V2.5-Dev — ROCmFP4_STRIX + grafted MTP head (GGUF)

	A 4-bit `ROCmFP4_STRIX` quant of [Kwaipilot/KAT-Coder-V2.5-Dev](https://huggingface.co/Kwaipilot/KAT-Coder-V2.5-Dev) with the model's MTP (multi-token prediction) head grafted back in at Q8_0, so it can self-speculate. Built and measured on an AMD Ryzen AI Max+ 395 (Strix Halo, gfx1151, 128 GB unified memory).

	19.0 GB — `KAT-Coder-V2.5-Dev-Q4_0_ROCMFP4_STRIX-MTP.gguf`

	## ⚠️ Requires a llama.cpp fork — stock llama.cpp cannot read this file

	`ROCmFP4` uses GGUF tensor types 100/101, which upstream llama.cpp does not know. You need [charlie12345/ROCmFPX](https://github.com/charlie12345/ROCmFPX). Stock `gguf-py` also can't parse it (use the fork's `gguf-py` for inspection).

	If you don't have that fork, this file is not for you — take a Q4_K_M or Q6_K build of the base model instead.

	## Serving

	```bash
	env HSA_OVERRIDE_GFX_VERSION=11.5.1 GGML_HIP_ENABLE_UNIFIED_MEMORY=1 \
	llama-server -m KAT-Coder-V2.5-Dev-Q4_0_ROCMFP4_STRIX-MTP.gguf \
	-ngl 999 -fa on -c 32768 --jinja --metrics \
	--spec-type draft-mtp --spec-draft-n-max 1 \
	--reasoning-format deepseek --reasoning-budget 0
	```

	- `--spec-draft-n-max 1` is what was tuned here. The MTP head predicts one token ahead; deeper drafting costs more in verification than it returns on this hardware.
	- `--reasoning-format deepseek --reasoning-budget 0` to suppress thinking. Do not use `--reasoning-format none` if you post-process the output: `none` tells llama.cpp not to parse think tags, so `</think>` is left inline in `content` and will corrupt anything that extracts code from the response. With `deepseek`, residue lands in `reasoning_content` instead. Expect ~72 chars of residue there that cannot be driven to zero.

	## Measured

	Same hardware, `--parallel 1`, greedy, thinking off (probe-verified), single runs.

	### Quality

	\| benchmark \| score \| notes \|
	\|---\|---:\|---\|
	\| GSM8K (n=500, 5-shot) \| 0.952 ±0.010 \| identical to the plain STRIX base — MTP costs no quality \|
	\| IFEval prompt_strict (n=500) \| 0.764 ±0.019 \| inst_strict 0.834 \|
	\| HumanEval (pass@1, 164) \| 0.939 \| \|
	\| HumanEval+ (pass@1, 164) \| 0.884 \| evalplus hardened tests \|

	### Speed

	\| \| plain STRIX \| this build (MTP) \|
	\|---\|---:\|---:\|
	\| decode @1k \| 67.6 t/s \| 92.5 t/s (+37%) \|
	\| decode @8k \| 63.8 t/s \| 87.1 t/s (+37%) \|

	## Caveats

	- MTP defeats the prompt cache (`spec-boundary-mismatch`), so this build suits single-shot work better than long multi-turn chat. If you need prompt-cache reuse, serve the plain STRIX quant without `--spec-type`.
	- Thinking-off numbers only. Reasoning-on was not benchmarked; the scores above are floors.
	- Single runs, greedy, pass@1. No repeats, so treat 1–2 point differences as noise.
	- HumanEval base is saturated at this capability level — HumanEval+ is the more discriminating number.
	- `--spec-type draft-mtp` works with `--parallel > 1` (per-slot draft state is maintained). But `n-max` should come down as concurrency rises, since the verify batch scales as `n-max × active slots`.

	## How it was built

	Quantized from a BF16 conversion of the base model with the base model's own imatrix, using the fork's `llama-quantize` at `Q4_0_ROCMFP4_STRIX` (a selective recipe: `attn_k`/`attn_v` get the quality layout, `attn_q`/`attn_output`/`ffn_*_exps` get the fast layout, `token_embd` gets Q6_K). The MTP head — which ships inside the base model rather than as a separate draft — was then grafted in at Q8_0.

	Credit to Kwaipilot for the base model, and to the ROCmFPX fork for the FP4 kernels.