Instructions to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-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 IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-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 IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0
# Run inference directly in the terminal:
llama cli -hf IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0
# Run inference directly in the terminal:
llama cli -hf IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF: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 IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0
# Run inference directly in the terminal:
./build/bin/llama cli -hf IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0

Use Docker

docker model run hf.co/IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0

Llama new
LM Studio
Jan

vLLM

How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-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": "IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0

Ollama
How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF with Ollama:
```
ollama run hf.co/IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0
```
Unsloth Desktop

How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF: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": "IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF with Docker Model Runner:
```
docker model run hf.co/IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0
```

Lemonade

How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0

Run and chat with the model

lemonade run user.KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF-Q4_0

List all available models

lemonade list

Hermes Agent

How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-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 IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF: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 IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF:Q4_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF: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 "IsValorum/KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF: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"

KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1

APEX Quantization Series by IsValorum

Family Cross-Links

Precision Tier	Standard Alignment (Base)
MiniPlus (14.75 GB)	KAT-Coder-V2.5-Dev-APEX-I-MiniPlus-V2.1-GGUF
NanoPlus (12.55 GB)	KAT-Coder-V2.5-Dev-APEX-I-NanoPlus-V2.1-GGUF

Architecture & Precision Highlights

Base Model: Kwaipilot/KAT-Coder-V2.5-Dev
Architecture Type: Sparse Mixture-of-Experts (Qwen 35B / 40 Layers / DeltaNet Hybrid)
Total Parameters: 34.7 Billion (approx. 3.4 Billion active per token)
Quantization Profile: APEX-I NanoPlus V2.1 (approx. 11.69 GiB / approx. 12.55 GB on web)
Audited Tensors: Exactly 733 verified tensors
Integrated Speculative Decoding: Bundled with standalone MTP adapter mtp-KAT-Coder-V2.5-Dev-Q4_0.gguf

Structural Group	Precision Tier	Technical Rationale
Recurrent Core & Norms	`F32` (`ssm_a`, `ssm_alpha`, `ssm_conv1d`, `ssm_dt`, `ssm_norm`, RMSNorms)	Shielded from quantization noise; protects continuous-state memory dynamics.
Recurrent Gates & Output	`Q4_K` (`ssm_beta`, `ssm_out`)	Linear-attention gate and output projections kept at 4-bit for AVX2 streaming.
MoE Routers	`F32`	`ffn_gate_inp` and shared gates locked to 32-bit floats to prevent expert collapse.
Shared Experts	`Q4_K`	Preserves fundamental token representations across all 40 layers.
Boundary Routed Experts	`Q3_K` (down) / `IQ3_XXS` (gate, up)	Layers 0-1 and 38-39 protected with higher precision codebooks.
Core Routed Experts	`IQ2_S` (gate) / `IQ2_S` + `IQ2_XXS` (up, layers 2-15 and 16-37)	Layers 2-37 compressed via importance matrix guidance.
Attention Projections	`Q4_K` (`attn_q/k/v`, `attn_qkv`) / `Q6_K` (`attn_output`)	Unified attention projection fidelity with an armored output projection.
Output Head	`Q6_K`	Highest-precision linear projection for vocabulary generation.

Recommended Generation Parameters

llama-server \
  -m KAT-Coder-V2.5-Dev.APEX-I-NanoPlus-V2.1.gguf \
  --draft-model mtp-KAT-Coder-V2.5-Dev-Q4_0.gguf \
  --draft-max 2 \
  --host 0.0.0.0 \
  --port 8080 \
  -c 32768 \
  -ngl 99 \
  --temp 0.6 \
  --top-p 0.95 \
  --top-k 40

Support & Community Compute Fund

All APEX-I quantizations are independently computed on private cloud clusters funded 100% out-of-pocket. If this release helps your workflow, consider supporting on Ko-fi. Every contribution directly sustains future cloud compute hours and supports my university studies.