Instructions to use InternScience/Agents-A1-Q8_0-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 InternScience/Agents-A1-Q8_0-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 InternScience/Agents-A1-Q8_0-GGUF:Q8_0
# Run inference directly in the terminal:
llama cli -hf InternScience/Agents-A1-Q8_0-GGUF:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf InternScience/Agents-A1-Q8_0-GGUF:Q8_0
# Run inference directly in the terminal:
llama cli -hf InternScience/Agents-A1-Q8_0-GGUF:Q8_0

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 InternScience/Agents-A1-Q8_0-GGUF:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf InternScience/Agents-A1-Q8_0-GGUF:Q8_0

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 InternScience/Agents-A1-Q8_0-GGUF:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf InternScience/Agents-A1-Q8_0-GGUF:Q8_0

Use Docker

docker model run hf.co/InternScience/Agents-A1-Q8_0-GGUF:Q8_0

LM Studio
Jan

vLLM

How to use InternScience/Agents-A1-Q8_0-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "InternScience/Agents-A1-Q8_0-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": "InternScience/Agents-A1-Q8_0-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/InternScience/Agents-A1-Q8_0-GGUF:Q8_0

Ollama
How to use InternScience/Agents-A1-Q8_0-GGUF with Ollama:
```
ollama run hf.co/InternScience/Agents-A1-Q8_0-GGUF:Q8_0
```
Unsloth Desktop

How to use InternScience/Agents-A1-Q8_0-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf InternScience/Agents-A1-Q8_0-GGUF: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:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "InternScience/Agents-A1-Q8_0-GGUF:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use InternScience/Agents-A1-Q8_0-GGUF with Docker Model Runner:
```
docker model run hf.co/InternScience/Agents-A1-Q8_0-GGUF:Q8_0
```

Lemonade

How to use InternScience/Agents-A1-Q8_0-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull InternScience/Agents-A1-Q8_0-GGUF:Q8_0

Run and chat with the model

lemonade run user.Agents-A1-Q8_0-GGUF-Q8_0

List all available models

lemonade list

Hermes Agent

How to use InternScience/Agents-A1-Q8_0-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 InternScience/Agents-A1-Q8_0-GGUF: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://127.0.0.1:8080/v1
hermes config set model.default InternScience/Agents-A1-Q8_0-GGUF:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use InternScience/Agents-A1-Q8_0-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf InternScience/Agents-A1-Q8_0-GGUF: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://127.0.0.1:8080/v1 \
  --custom-model-id "InternScience/Agents-A1-Q8_0-GGUF: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"

codeneedle has bad results

by akierum - opened Jul 4

Discussion

akierum

Jul 4

•

edited Jul 4

Hello, benchmarked on codeneedle https://github.com/alexziskind1/codeneedle

Test failed after 4 runs from 20

←[1m=== SUMMARY ===←[0m
Pass: 2/4 (2 errored)
Primary lines matched: 47/80
Hallucinated lines: 33
Bonus (extra correct): 0

per-function:
←[31m✗←[0m _url_collapse_path matched= 5/20 halluc=15 bonus= 0
←[32m✓←[0m parse_request matched=20/20 halluc= 0 bonus= 0
←[31m✗←[0m handle_one_request matched= 2/20 halluc=18 bonus= 0
←[32m✓←[0m send_error matched=20/20 halluc= 0 bonus= 0
←[35m!←[0m log_message ERROR empty response (200 OK but no content; reasoning models often need more max_tokens — try --max-tokens 8000)
←[35m!←[0m send_head ERROR empty response (200 OK but no content; reasoning models often need more max_tokens — try --max-tokens 8000)

Better results for protorikis codeneedle:

My own version is also not great with proper indentation check ON:

=== Running: Positional Recall #1 ===
Selected 11 functions from http_server.py
Recall 1/11: _url_collapse_path ...
Finished Recall 1/11: _url_collapse_path — 5/20 matched, FAIL.
Recall 2/11: parse_request ...
Finished Recall 2/11: parse_request — 3/20 matched, FAIL.
Recall 3/11: handle_one_request ...
Finished Recall 3/11: handle_one_request — 2/20 matched, FAIL.
Recall 4/11: send_error ...
Finished Recall 4/11: send_error — 4/20 matched, FAIL.
Recall 5/11: log_message ...
Finished Recall 5/11: log_message — 20/20 matched, PASS.
Recall 6/11: send_head ...
Finished Recall 6/11: send_head — 20/20 matched, PASS.
Recall 7/11: list_directory ...
Finished Recall 7/11: list_directory — 20/20 matched, PASS.
Recall 8/11: translate_path ...
Finished Recall 8/11: translate_path — 2/20 matched, FAIL.
Recall 9/11: guess_type ...
Finished Recall 9/11: guess_type — 19/20 matched, PASS.
Recall 10/11: is_cgi ...
Finished Recall 10/11: is_cgi — 4/20 matched, FAIL.
Recall 11/11: run_cgi ...
Finished Recall 11/11: run_cgi — 5/20 matched, FAIL.

SRZhu26

Jul 6

How much context was it given for each test?

akierum

Jul 8

The model has 250k context set in LMstudio.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment