Skip to main content

cloooooo — personal LLM client, prompt/context/thinking interface over local Ollama

Project description

clovis

Client Python personnel pour un LLM local via Ollama. Interface ultra-simple : prompt, negative_prompt, thinking, context.

Install

pip install clovis

Usage

from clovis import cloooooo

ai = cloooooo()

# Appel direct
print(ai("Explique les trous noirs"))

# Avec options
print(ai(
    "Génère un poème sur la mer",
    negative_prompt="pas de rimes",
    thinking=True,
    context="Tu es un poète du 19e siècle.",
))

# Streaming token par token
for token in ai.stream("Raconte une histoire courte"):
    print(token, end="", flush=True)

# Conversation avec mémoire
conv = ai.conversation(context="Tu es un expert en finance.")
conv("Explique le CAPM")
conv("Et ses limites ?")  # se souvient de la réponse précédente

CLI

clovis "Explique les trous noirs"           # question directe
clovis "Génère un poème" --no "sans rimes"  # avec negative prompt
clovis "Résous ce problème" --think          # mode réflexion
clovis repl                                  # conversation interactive
clovis serve --port 8000                     # démarre le serveur API

API server

clovis serve --port 8000 --key sk-montoken

Requête :

curl -X POST http://localhost:8000/ia \
  -H "Authorization: Bearer sk-montoken" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Bonjour !", "thinking": false}'

Réponse : {"response": "Bonjour ! Comment puis-je t'aider ?"}

Streaming : ajouter "stream": true → réponse en text/plain token par token.

Config

export CLOVIS_MODEL="qwen3-72b-q5km"        # modèle Ollama
export CLOVIS_OLLAMA_URL="http://localhost:11434"
export CLOVIS_API_KEY="sk-..."              # clé API pour le serveur

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

clovis-0.4.0.tar.gz (12.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

clovis-0.4.0-py3-none-any.whl (10.3 kB view details)

Uploaded Python 3

File details

Details for the file clovis-0.4.0.tar.gz.

File metadata

  • Download URL: clovis-0.4.0.tar.gz
  • Upload date:
  • Size: 12.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for clovis-0.4.0.tar.gz
Algorithm Hash digest
SHA256 56db332b529106d129e3e5721e1c36f67dc18be20fb38f8e607f9ffadd30bc9d
MD5 5e80f114336e2e5f21b650052ce9d8fe
BLAKE2b-256 d178fee7c77e8ae78e59d656bff31f21520c431cf3129ab1b8e4cd709ff6c758

See more details on using hashes here.

File details

Details for the file clovis-0.4.0-py3-none-any.whl.

File metadata

  • Download URL: clovis-0.4.0-py3-none-any.whl
  • Upload date:
  • Size: 10.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for clovis-0.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 33a5dc188f2027821307011c356c50aa9445b80292324e6a49e6a91f10476816
MD5 204449bef918c604248b0dbdb58016f0
BLAKE2b-256 b6c40d6f7dc1587316d99784e6210590de15f7a7bda054a053f2a53ab29f534c

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page