Covalent Blueprints are pre-configured applications for Covalent. Each blueprint is runnable both on its own and as a component in another workflow. See the catalogue below for a list of available blueprints.
Example: Deploy a Llama 3 chatbot backend
Run a Llama3 chatbot on H100 GPUs in just a few lines.
from covalent_blueprints import store_secret, save_api_key
from covalent_blueprints_ai import llama_chatbot
# Set credentials
save_api_key("<covalent-cloud-api-key>")
store_secret(name="HF_TOKEN", value="<huggingface-write-token>")
# Initialize a blueprint
bp = llama_chatbot(model_name="meta-llama/Meta-Llama-3-70B-Instruct")
# Customize compute resources (e.g. 2x H100 GPUs)
bp.executors.service_executor.gpu_type = "h100"
bp.executors.service_executor.num_gpus = 2
bp.executors.service_executor.memory = "240GB"
# Run the blueprint
llama_client = bp.run()
The llama_chatbot blueprint returns a Python client for the deployed service.
llama_client.generate(prompt="How are you feeling?", max_new_tokens=100)
How are you feeling? How are you doing?
I am feeling well, thank you for asking. I am a machine learning model, so I don't have emotions or feelings in the way that humans do.
llama_client.generate_message(
messages=[
{"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
{"role": "user", "content": "Who are you?"},
]
)
{'role': 'assistant', 'content': "Arrrr, me hearty! Me be Captain Chatterbeard, the scurviest chatbot to ever sail the seven seas o' conversation! Me be here to swab yer decks with me witty banter, me treasure trove o' knowledge, and me trusty cutlass o' clever responses! So hoist the colors, me matey, and set course fer a swashbucklin' good time! What be bringin' ye to these fair waters?"}
Release compute resources with a single line.
llama_client.teardown()
Blueprints catalogue
👉 Each link below points to an example notebook.
pip install -U covalent-blueprints-ai
| Blueprint | Description |
|---|---|
| Image Generator | Deploy a text-to-image generator service. |
| Llama Chatbot | Deploy a chatbot backend using a Llama-like model. |
| LoRA fine tuning | Fine tune and deploy an LLM as a Covalent service. |
| vLLM | Deploy an LLM using vLLM on Covalent Cloud. |
| NVIDIA Llama RAG | Deploy a retrieval-augmented generation (RAG) pipeline using multiple NVIDIA NIMs. |
More coming soon...
Contributing
Public contributions will soon be open! In the meantime, please reach out on Slack to contribute a blueprint.
1 maintainer
Metadata
1 maintainer
Release files for covalent-blueprints 0.6.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| covalent_blueprints-0.6.1.tar.gz | 35.8 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| covalent_blueprints-0.6.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 75.7 kB
Release files / covalent_blueprints-0.6.1.tar.gz
| Download URL | covalent_blueprints-0.6.1.tar.gz |
|---|---|
| Size | 35.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f794f2e6e335c3f5453c9e26c96e08a7c9fb5dd8c39226c243369407bc3e96e2
|
|
BLAKE2b-256 checksum How to use checksums |
15bc80c2c556ae45e1c45701216a3bac9f4b5a12303de446b1ba2c37a281a40b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.10.15
|
Release files / covalent_blueprints-0.6.1-py3-none-any.whl
| Download URL | covalent_blueprints-0.6.1-py3-none-any.whl |
|---|---|
| Size | 39.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a992486dd2ed61c06665f795cd6432f632c8d1a8fd1b09b5ecd55cf43612db1f
|
|
BLAKE2b-256 checksum How to use checksums |
a729ab4ad17be0298881d52deb596ec04183c1073c2439bac95fcddb529486fd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.1 CPython/3.10.15
|
Security for covalent-blueprints 0.6.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
