$ git clone https://github.com/instill-ai/instill-core
$ cd instill-core
$ make all
Complex Data Cleaning
80% of enterprise data is unstructured in unusable format and difficult to clean.
Broken or Rigid Data Flow
Data flows for AI can be very complex, requiring rich integrations and and customized automation.
Hallucinations in answers
Stuck with LLMs that produce inaccurate, irrelevant or even fabricated answers.
Struggling to Scale
You're bogged down by infrastructure maintenance work that has little business impact.
Expensive
It's tough to control budgets and get quick results with complex AI systems.
Fragmented Tools
Your data, models, workflows and operations are fragmented across a galaxy of tools.
Instill Catalog easily converts documents, images, audio, and video into a unified format. It simplifies data cleaning, reduces errors, automates updates, and ensures your data is RAG-ready for AI applications. With built-in data lineage support, you can track data origins, transformations, and flows, ensuring transparency and maintaining data integrity.
$ curl -X GET "https://api.instill.tech/v1alpha/namespaces/instill-ai/catalogs/product-knwl/file-catalog?fileId=intro.pdf" \
-H "Authorization: Bearer instill_sk_ncciFtg4..."
# Response
{
"metadata": {
"fileUid": "3a473eaf-2ca9-4c9c-b252-f43522bc4e91",
"fileId": "intro.pdf",
"fileUploadTime": "2024-08-15T02:18:51.056874Z",
"fileProcessStatus": "FILE_PROCESS_STATUS_COMPLETED",
...
},
"text": {
"pipelineIds": [
"preset/indexing-split-markdown@v2.0.0",
"preset/indexing-embed@v1.1.0"
],
"transformedContent": "💾 Instill Artifact orchestrates unstructured data to transform documents, images, audio, and video into Instill Catalog - a unified AI-ready format.",
"transformedContentChunkNum": 9,
...
},
"chunks":[
{
"uid": "00666d84-9bfe-4c5c-b4ff-4b65faad6d93",
"startPos": 10,
"endPos": 433,
"content": "# 💾 Instill Artifact orchestrates...",
"embedding": [0.0041343034, -0.024816118,...],
...
},
...
]
}
Retrieve relevant results grounded in your data via simple APIs provided by Instill Catalog. Ideal for developers building intelligent search and Q&A services, such as AI assistants, with no deep technical expertise in LLM or RAG required.
$ curl -X POST "https://api.instill.tech/v1alpha/namespaces/instill-ai/catalogs/product-knwl/chunks/similarity" \
-H "Authorization: Bearer instill_sk_ncciFtg4..." \
-d '{
"textPrompt": "what is instill core?",
"topK": 5
}'
# Response
{
"similarChunks": [
{
"chunkUid": "ba30f524-889c-4dc7-82a2-33a8f7be2d47",
"similarityScore": 0.95,
"textContent": "Instill Core is a full-stack AI solution to accerlerate AI development...",
"sourceFile": "core.txt"
},
{
"chunkUid": "757ab6d9-e5b4-482e-8017-5582b578e57a",
"similarityScore": 0.89,
"textContent": "Transform unstructured data into Instill catalog to be AI-ready...",
"sourceFile": "intro.pdf"
},
...
]
}
Flexibly use state-of-the-art models across different vendors, or you can run any open-source AI models, automatically scale up and down, ensuring reliable compute resources without manual maintenance. Our high-performance platform serves your AI models with ease. From Setup to Scaling, We've Got Infra Covered.
Model Vendors
Instill Model
Fix broken data flows and improve data quality to your AI applications. Create DAGs of dependencies, connect third-party data sources, AI models, operations, and applications. Automate ETL processes to manage and orchestrate the entire pipeline in a single interface, powering your applications and business efficiently.
Forget about gluing different tools together. Our platform integrates seamlessly with your systems, making scaling your AI applications effortless. It streamlines data handling, enhances AI accuracy, continuously optimizes through monitoring, and effectively manages infrastructure. This allows you to expand your AI applications efficiently and cost-effectively. Enjoy native compatibility at every stage for hassle-free maintenance.
Cloud-native
Fully managed on your choice of public cloud, Bring Your Own Cloud (BYOC), or on-premises.
Security & Privacy
Control your data securely with TLS encryption and strict retention policies.
10X
Faster to develop and ship your AI applications
30%
Capacity boost to break down data and AI team silos
$2M
Saved on R&D budget for more efficient resource use
Find out how your business can benefit from AI
AI infrastructure for Enterprise