Text Generation
PEFT
Safetensors
GGUF
PyTorch
English
afm-d
afm-dd
decision
system-1
minicpm
minicpm5
lora
semif
conversational
Instructions to use ariacompute/afm-dd with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use ariacompute/afm-dd with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("openbmb/MiniCPM5-2B") model = PeftModel.from_pretrained(base_model, "ariacompute/afm-dd") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use ariacompute/afm-dd 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 ariacompute/afm-dd:Q8_0 # Run inference directly in the terminal: llama cli -hf ariacompute/afm-dd:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf ariacompute/afm-dd:Q8_0 # Run inference directly in the terminal: llama cli -hf ariacompute/afm-dd:Q8_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 ariacompute/afm-dd:Q8_0 # Run inference directly in the terminal: ./build/bin/llama cli -hf ariacompute/afm-dd:Q8_0
Use Docker
docker model run hf.co/ariacompute/afm-dd:Q8_0
- Llama new
- LM Studio
- Jan
- vLLM
How to use ariacompute/afm-dd with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ariacompute/afm-dd" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ariacompute/afm-dd", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/ariacompute/afm-dd:Q8_0
- Ollama
How to use ariacompute/afm-dd with Ollama:
ollama run hf.co/ariacompute/afm-dd:Q8_0
- Unsloth Desktop
- Pi
How to use ariacompute/afm-dd with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf ariacompute/afm-dd:Q8_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:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "ariacompute/afm-dd:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use ariacompute/afm-dd with Docker Model Runner:
docker model run hf.co/ariacompute/afm-dd:Q8_0
- Lemonade
How to use ariacompute/afm-dd with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull ariacompute/afm-dd:Q8_0
Run and chat with the model
lemonade run user.afm-dd-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use ariacompute/afm-dd with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf ariacompute/afm-dd:Q8_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://127.0.0.1:9931/v1 hermes config set model.default ariacompute/afm-dd:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use ariacompute/afm-dd with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf ariacompute/afm-dd:Q8_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://127.0.0.1:9931/v1 \ --custom-model-id "ariacompute/afm-dd:Q8_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"
Update README.md
Browse files
README.md
CHANGED
|
@@ -71,21 +71,21 @@ Load the adapter on the pinned MiniCPM5-2B revision with `peft` + SemIf / AFM-D
|
|
| 71 |
|
| 72 |
## How to use
|
| 73 |
|
| 74 |
-
Load through **
|
| 75 |
|
| 76 |
-
###
|
| 77 |
|
| 78 |
-
|
| 79 |
|
| 80 |
```bash
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
```
|
| 87 |
|
| 88 |
-
**System One** (`POST /v1/systemone`
|
| 89 |
|
| 90 |
```bash
|
| 91 |
curl -s http://127.0.0.1:8011/v1/systemone \
|
|
@@ -107,22 +107,10 @@ curl -s http://127.0.0.1:8011/v1/systemone \
|
|
| 107 |
|
| 108 |
| `type` | `criteria` |
|
| 109 |
| --- | --- |
|
| 110 |
-
| `choice` | object: option name → description
|
| 111 |
| `score` | array of 2–10 ordered level strings |
|
| 112 |
| `noul` | object with `true` / `false` (optional) |
|
| 113 |
|
| 114 |
-
### Aria Engine
|
| 115 |
-
|
| 116 |
-
Native AFM runtime: CLI, `POST /v1/systemone`, FFI, and language SDKs. Docs: [engine README](https://github.com/ariacompute/engine). Decoder System One default port **8011**.
|
| 117 |
-
|
| 118 |
-
```bash
|
| 119 |
-
aria-engine setup
|
| 120 |
-
aria-engine download afm-dd # → ~/.ariacompute/models/afm-dd
|
| 121 |
-
|
| 122 |
-
aria-engine decide --track decoder --model-name afm-dd --file record.json
|
| 123 |
-
aria-engine serve --track decoder --model-name afm-dd --bind 127.0.0.1:8011
|
| 124 |
-
```
|
| 125 |
-
|
| 126 |
**Python SDK** (`pip install aria-engine`; needs `libaria-engine_ffi` or `ARIA_FFI_LIB`):
|
| 127 |
|
| 128 |
```python
|
|
@@ -150,19 +138,25 @@ Also: TypeScript `@ariacompute/engine-ts`, Rust `ariacompute-engine`, Go / Flutt
|
|
| 150 |
|
| 151 |
### Ollaya
|
| 152 |
|
| 153 |
-
[Ollaya](https://ollaya.dev) pulls
|
| 154 |
|
| 155 |
```bash
|
|
|
|
| 156 |
curl -fsSL https://raw.githubusercontent.com/ariacompute/ollaya/main/scripts/install.sh \
|
| 157 |
| OLLAYA_REPO=ariacompute/ollaya sh
|
|
|
|
|
|
|
|
|
|
| 158 |
|
| 159 |
# weights: HF (ariacompute/afm-dd) or ModelScope (AriaCompute/afm-dd)
|
| 160 |
-
# OLLAYA_HUB=huggingface|modelscope|auto
|
| 161 |
ollaya pull afm-dd
|
| 162 |
ollaya run afm-dd --preset triage \
|
| 163 |
"Third time this year you've double-charged me. Refund it today or I'm cancelling."
|
| 164 |
```
|
| 165 |
|
|
|
|
|
|
|
| 166 |
HTTP (daemon default `http://localhost:11435`):
|
| 167 |
|
| 168 |
```bash
|
|
|
|
| 71 |
|
| 72 |
## How to use
|
| 73 |
|
| 74 |
+
Load through **Aria Engine** (native AFM runtime) or **[Ollaya](https://ollaya.dev)** (Ollama-style local decision daemon).
|
| 75 |
|
| 76 |
+
### Aria Engine
|
| 77 |
|
| 78 |
+
Native AFM runtime: CLI, `POST /v1/systemone`, FFI, and language SDKs. Docs: [engine README](https://github.com/ariacompute/engine). Decoder System One default port **8011**.
|
| 79 |
|
| 80 |
```bash
|
| 81 |
+
aria-engine setup
|
| 82 |
+
aria-engine download afm-dd # → ~/.ariacompute/models/afm-dd
|
| 83 |
+
|
| 84 |
+
aria-engine decide --track decoder --model-name afm-dd --file record.json
|
| 85 |
+
aria-engine serve --track decoder --model-name afm-dd --bind 127.0.0.1:8011
|
| 86 |
```
|
| 87 |
|
| 88 |
+
**System One** body (`POST /v1/systemone` or `decide`):
|
| 89 |
|
| 90 |
```bash
|
| 91 |
curl -s http://127.0.0.1:8011/v1/systemone \
|
|
|
|
| 107 |
|
| 108 |
| `type` | `criteria` |
|
| 109 |
| --- | --- |
|
| 110 |
+
| `choice` | object: option name → description |
|
| 111 |
| `score` | array of 2–10 ordered level strings |
|
| 112 |
| `noul` | object with `true` / `false` (optional) |
|
| 113 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 114 |
**Python SDK** (`pip install aria-engine`; needs `libaria-engine_ffi` or `ARIA_FFI_LIB`):
|
| 115 |
|
| 116 |
```python
|
|
|
|
| 138 |
|
| 139 |
### Ollaya
|
| 140 |
|
| 141 |
+
[Ollaya](https://ollaya.dev) pulls AFM-D by name and serves TypeSafe-compatible `/v1/systemone`. Use the **AFM-D-enabled** builds from [ariacompute/ollaya releases](https://github.com/ariacompute/ollaya/releases) (not the default `ollaya-dev/ollaya` channel). Weights stay on Hugging Face / ModelScope (Ollaya does not re-host).
|
| 142 |
|
| 143 |
```bash
|
| 144 |
+
# install from https://github.com/ariacompute/ollaya/releases (latest AFM-D build)
|
| 145 |
curl -fsSL https://raw.githubusercontent.com/ariacompute/ollaya/main/scripts/install.sh \
|
| 146 |
| OLLAYA_REPO=ariacompute/ollaya sh
|
| 147 |
+
# pin a release: OLLAYA_REPO=ariacompute/ollaya OLLAYA_VERSION=0.7.5+afm-d.1.0.0
|
| 148 |
+
# Windows (PowerShell):
|
| 149 |
+
# $env:OLLAYA_REPO='ariacompute/ollaya'; irm https://raw.githubusercontent.com/ariacompute/ollaya/main/scripts/install.ps1 | iex
|
| 150 |
|
| 151 |
# weights: HF (ariacompute/afm-dd) or ModelScope (AriaCompute/afm-dd)
|
| 152 |
+
# OLLAYA_HUB=huggingface|modelscope|auto (auto → ModelScope when LANG looks Chinese)
|
| 153 |
ollaya pull afm-dd
|
| 154 |
ollaya run afm-dd --preset triage \
|
| 155 |
"Third time this year you've double-charged me. Refund it today or I'm cancelling."
|
| 156 |
```
|
| 157 |
|
| 158 |
+
Or download a platform asset from the [releases page](https://github.com/ariacompute/ollaya/releases) (e.g. `ollaya-linux-amd64.tar.zst`, `ollaya-darwin-arm64.tgz`, `ollaya-windows-amd64.zip`), unpack, put `ollaya` on `PATH`, then `pull` / `run` as above.
|
| 159 |
+
|
| 160 |
HTTP (daemon default `http://localhost:11435`):
|
| 161 |
|
| 162 |
```bash
|