Instructions to use kanaton/webcoder-3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use kanaton/webcoder-3b with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use kanaton/webcoder-3b 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 kanaton/webcoder-3b:Q8_0 # Run inference directly in the terminal: llama cli -hf kanaton/webcoder-3b:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf kanaton/webcoder-3b:Q8_0 # Run inference directly in the terminal: llama cli -hf kanaton/webcoder-3b: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 kanaton/webcoder-3b:Q8_0 # Run inference directly in the terminal: ./llama-cli -hf kanaton/webcoder-3b: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 kanaton/webcoder-3b:Q8_0 # Run inference directly in the terminal: ./build/bin/llama-cli -hf kanaton/webcoder-3b:Q8_0
Use Docker
docker model run hf.co/kanaton/webcoder-3b:Q8_0
- LM Studio
- Jan
- vLLM
How to use kanaton/webcoder-3b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "kanaton/webcoder-3b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "kanaton/webcoder-3b", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/kanaton/webcoder-3b:Q8_0
- Ollama
How to use kanaton/webcoder-3b with Ollama:
ollama run hf.co/kanaton/webcoder-3b:Q8_0
- Unsloth Desktop
- Pi
How to use kanaton/webcoder-3b with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf kanaton/webcoder-3b: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": "kanaton/webcoder-3b:Q8_0" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use kanaton/webcoder-3b with Docker Model Runner:
docker model run hf.co/kanaton/webcoder-3b:Q8_0
- Lemonade
How to use kanaton/webcoder-3b with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull kanaton/webcoder-3b:Q8_0
Run and chat with the model
lemonade run user.webcoder-3b-Q8_0
List all available models
lemonade list
- Hermes Agent
How to use kanaton/webcoder-3b with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf kanaton/webcoder-3b: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 kanaton/webcoder-3b:Q8_0
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use kanaton/webcoder-3b with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf kanaton/webcoder-3b: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 "kanaton/webcoder-3b: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"
WebCoder 3B (webcoder-3b)
Assistant local de dรฉveloppement Web (PHP, HTML, CSS, SQL, JavaScript), adaptรฉ par LoRA, fusionnรฉ puis quantifiรฉ en GGUF. Conรงu pour tourner sur CPU, sans GPU.
- Auteur / identitรฉ : WebCoder, crรฉรฉ par Pierre Jean Duvivier (l'identitรฉ est apprise et vรฉrifiรฉe).
- Base :
Qwen/Qwen2.5-Coder-3B-Instruct@488639f1ff808d1d3d0ba301aef8c11461451ec5(Apache-2.0). - Format : GGUF (
Q8_0), prรชt pourllama.cpp/llama-server. - Fichier :
webcoder-3b-q8_0.gguf(~3,06 Gio).
Description
webcoder-3b est un petit modรจle spรฉcialisรฉ pour รฉcrire et corriger du code Web.
Il combine :
- une base Qwen2.5-Coder-3B-Instruct (mรชme famille que le 1,5B utilisรฉ auparavant) ;
- un LoRA ยซ seed ยป (rang 8,
q_proj/v_proj) appris sur des corpus d'identitรฉ, de chat et de worker (`format de blocsfile:CHEMINpour รฉditer des fichiers) ; - une fusion dans les poids de base, puis une quantification Q8_0 en GGUF.
Le modรจle rรฉpond en franรงais par dรฉfaut, connaรฎt son identitรฉ (ยซ WebCoder ยป), et sait produire du code Web dans le format worker (un bloc par fichier, contenu complet).
Utilisation (llama.cpp / llama-server, CPU)
llama-server --model webcoder-3b-q8_0.gguf \
--ctx-size 4096 --parallel 1 --n-gpu-layers 0 \
--threads 6 --host 127.0.0.1 --port 8081
Puis, via l'API OpenAI-compatible :
curl http://127.0.0.1:8081/v1/chat/completions -H "Content-Type: application/json" -d '{
"model": "webcoder-3b",
"messages": [{"role":"user","content":"Ecris une fonction PHP qui inverse une chaine."}],
"max_tokens": 256, "temperature": 0.2
}'
Le format worker attend un bloc par fichier modifiรฉ :
```file:src/exemple.php
<?php
...contenu complet du fichier...
```
Entraรฎnement (rรฉsumรฉ)
| รlรฉment | Valeur |
|---|---|
| Base | Qwen2.5-Coder-3B-Instruct @ 488639f1โฆ |
| Mรฉthode | LoRA (PEFT), rang 8, alpha 16, dropout 0.05โ0.1, cibles q_proj,v_proj |
| Corpus seed | identitรฉ + chat + worker (chat_corpus + t28g_corpus, 317 ex.) |
| Pas | 150 microsteps, LR 5e-5, sรฉlection sur validation |
| Prรฉcision | base FP32 (CPU, sans GPU) |
| Export | fusion (PEFT merge_and_unload) โ GGUF f16 โ llama-quantize Q8_0 |
| Dtype servi | Q8_0 (~3,06 Gio) |
Entraรฎnement CPU uniquement (politique CPU-first du projet).
รvaluation (banc interne ยซ apps PHP ยป, format worker)
Rรฉussite au 1er essai sur des tรขches PHP from-scratch (rรฉserve tenue ร l'รฉcart) :
| Famille | Score |
|---|---|
| Fonctions courtes (facile) | 8/8 |
Multi-fonctions / รฉtat (z) |
5/5 |
Divers (y) |
4/5 |
Programmes multi-fichiers (p) |
2/4 |
Identitรฉ (identity_check) |
OK |
| Chat (4 questions) | 4/4 |
Comparรฉ ร un modรจle 1,5B (mรชme recette), la 3B est nettement meilleure sur les tรขches dures (la capacitรฉ รฉtait le facteur limitant).
Limites
- Pas agentique : pas de tool-calling structurรฉ ; ร utiliser comme assistant de code / worker derriรจre un orchestrateur dรฉterministe.
- Applications Web complรจtes depuis zรฉro (ex. un jeu du pendu web fonctionnel) : encore hors de portรฉe (รฉchec mesurรฉ ; persistance d'รฉtat en session).
- Le code gรฉnรฉrรฉ n'est pas garanti sรปr ; revue humaine requise pour paiements, authentification, migrations.
- Qualitรฉ hors du pรฉrimรจtre Web (PHP/HTML/CSS/SQL/JS) non mesurรฉe.
- Infรฉrence CPU lente au premier message (prรฉfill), rapide ensuite (cache KV).
Licence
Modรจle de base sous Apache-2.0. Les adaptations et l'export conservent cette licence.
Voir le LICENSE du modรจle de base.
Citation / attribution
WebCoder 3B (webcoder-3b) โ Pierre Jean Duvivier.
Base : Qwen2.5-Coder-3B-Instruct (Alibaba Cloud, Apache-2.0).
- Downloads last month
- 11
8-bit