Instructions to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

Use Docker

docker model run hf.co/osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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": "osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

Ollama
How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF with Ollama:
```
ollama run hf.co/osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M
```
Unsloth Desktop

How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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": "osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF with Docker Model Runner:
```
docker model run hf.co/osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M
```

Lemonade

How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.LFM2.5-8B-A1B-ThinkingCap-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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 osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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 "osk-arr00/LFM2.5-8B-A1B-ThinkingCap-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"

🧠 LFM 2.5 8B ThinkingCap — GGUF & ROCmFPX Suite

Abliterated • SFT + DPO ThinkingCap • APEX Tensor-Mapped • Native ROCmFP4 for AMD Strix Halo (gfx1151)

📌 Model Overview

LFM 2.5 8B ThinkingCap is a distilled, reasoning-calibrated, and abliterated hybrid architecture combining 36 layers of 1D short-convolutions with a 32-expert MoE backbone (~1.2B active parameters per token out of 8.47B total).

Trained using full Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO), this model enforces the ThinkingCap contract:

Focused internal reasoning enclosed strictly within [Start thinking] ... [End thinking].
Clean, structured, and deterministic answers in the deliverable block.
High fidelity and fluency in English and Spanish.

⚡ Quantization & APEX Matrix

This repository provides standard GGUF quantizations alongside APEX mixed-precision and native ROCmFP4 quants specifically tailored for the AMD Strix Halo (RDNA 3.5 / gfx1151 / Radeon 8050S / Radeon 8060S) platform with unified LPDDR5X memory.

Available GGUF Files

File Name	Size	Type / Description	Target Architecture
`LFM2.5-8B-A1B-ThinkingCap-ROCMFPX-APEX.gguf`	4.78 GB	Definitive ROCmFPX APEX: Routed experts in `Q4_0_ROCMFP4`, attention in `Q6_K`, shared experts/conv in `Q8_0`	AMD Strix Halo (`gfx1151`) ROCmFPX
`LFM2.5-8B-A1B-ThinkingCap-APEX.gguf`	4.63 GB	Universal APEX: Routed experts in `Q4_K_M` + `apex.txt` + `iMatrix`, attention in `Q6_K`, conv in `Q8_0`	llama.cpp / Vulkan / CPU / CUDA
`LFM2.5-8B-A1B-ThinkingCap-APEX-Q6.gguf`	6.60 GB	APEX Q6: Maximum reasoning retention with moderate compression	High-precision inference
`LFM2.5-8B-A1B-ThinkingCap-Q8_0.gguf`	8.40 GB	Full 8-bit reference	Reference evaluation
`LFM2.5-8B-A1B-ThinkingCap-Q6_K.gguf`	6.50 GB	Standard Q6_K	General purpose
`LFM2.5-8B-A1B-ThinkingCap-BF16.gguf`	16.0 GB	Unquantized BF16 Base	Research / Conversion base
`LFM2.5-8B-A1B-DSpark-Q4_K_M.gguf`	191 MB	DSpark Speculative Drafter (Q4_K_M)	Speculative decoding drafter
`LFM2.5-8B-A1B-DSpark-BF16.gguf`	633 MB	DSpark Drafter (BF16)	Speculative decoding base

📊 Measured Benchmarks (AMD Radeon 8050S / Strix Halo gfx1151)

All benchmarks measured locally on AMD Ryzen AI Max 385 APU with 64 GB LPDDR5X Unified RAM.

1. Prefill & Generation Context Ladder (`llama-bench`)

Metric	Context Size	GPU Performance
Short Prefill (`pp512`)	512 tokens	3,061.1 tokens/sec
Medium Prefill (`pp1024`)	1,024 tokens	3,165.7 tokens/sec
Long Prefill (`pp4096`)	4,096 tokens	3,564.4 tokens/sec
Deep Prefill (`pp16384`)	16,384 tokens	2,233.8 tokens/sec
Decode / Generation (Single-Stream)	128 – 4,096 tokens	115.8 – 148.3 tokens/sec

2. Multi-Slot Continuous Batching Scaling (256K Global Context)

Concurrency	Per-User Decode Speed	Aggregate System Throughput	Scaling Factor
1 Slot	148.3 tokens/sec	121.5 tokens/sec	1.00× (Baseline)
2 Slots	75.2 tokens/sec	136.4 tokens/sec	1.12×
4 Slots	64.8 tokens/sec	211.8 tokens/sec	1.74×

🎯 Recommended Sampling Parameters (DRY Sampler Sweetspot)

For optimal reasoning output and zero repetitive loops, use the DRY (Don't Repeat Yourself) sampler configuration instead of static repetition penalty:

{
  "temperature": 0.2,
  "top_p": 0.95,
  "dry_multiplier": 0.8,
  "dry_base": 1.75,
  "dry_allowed_length": 2,
  "max_tokens": -1
}

🚀 Quickstart & Deployment

Run with `llama-cli` (ROCm / HIP for gfx1151)

export HSA_OVERRIDE_GFX_VERSION=11.5.1
export GGML_HIP_ENABLE_UNIFIED_MEMORY=1

llama-cli \
  -m LFM2.5-8B-A1B-ThinkingCap-ROCMFPX-APEX.gguf \
  -dev ROCm0 \
  -ngl 999 \
  -fa on \
  -ctk q8_0 -ctv q8_0 \
  -b 8192 -ub 2048 \
  -c 32768 \
  --dry-multiplier 0.8 --dry-base 1.75 --dry-allowed-length 2 \
  --temp 0.2 \
  --jinja \
  -p "<|im_start|>system\nEres un asistente experto.\n<|im_end|>\n<|im_start|>user\nHola, ¿cuál es tu arquitectura?\n<|im_end|>\n<|im_start|>assistant\n"

Production Server (Docker)

docker run -d \
  --name strix-halo-server-fast \
  --restart unless-stopped \
  --device=/dev/dri \
  --group-add 983 --group-add 987 \
  -p 8080:8080 \
  -v /path/to/models:/models:ro \
  strix-halo-llamacpp:ultimate-vulkan \
  llama-server \
    -m /models/LFM2.5-8B-A1B-ThinkingCap-ROCMFPX-APEX.gguf \
    -ngl 999 \
    --host 0.0.0.0 \
    --port 8080 \
    -fa 1 \
    -ctk q8_0 -ctv q8_0 \
    -b 16384 -ub 2048 \
    -np 4 -c 262144 \
    --jinja \
    --metrics

📄 License & Credits

Base Weights: Based on Liquid AI LFM 2.5 8B A1B and huihui-ai abliterated.
Fine-Tuning: ThinkingCap SFT & DPO training by osk-arr00.
ROCmFPX Kernels: Built with ROCmFP4 custom Wave32 ISA kernels for AMD RDNA 3.5 (gfx1151).

Downloads last month: 311

GGUF

Model size

8B params

Architecture

lfm2moe

Hardware compatibility

4-bit

6-bit

8-bit

16-bit

View +3 variants

Model tree for osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF

Base model

osk-arr00/LFM2.5-8B-A1B-ThinkingCap

Quantized

(1)

this model

🧠 LFM 2.5 8B ThinkingCap — GGUF & ROCmFPX Suite

📌 Model Overview

⚡ Quantization & APEX Matrix

Available GGUF Files

📊 Measured Benchmarks (AMD Radeon 8050S / Strix Halo gfx1151)

1. Prefill & Generation Context Ladder (llama-bench)

2. Multi-Slot Continuous Batching Scaling (256K Global Context)

🎯 Recommended Sampling Parameters (DRY Sampler Sweetspot)

🚀 Quickstart & Deployment

Run with llama-cli (ROCm / HIP for gfx1151)

Production Server (Docker)

📄 License & Credits

Model tree for osk-arr00/LFM2.5-8B-A1B-ThinkingCap-GGUF

1. Prefill & Generation Context Ladder (`llama-bench`)

Run with `llama-cli` (ROCm / HIP for gfx1151)