Knowledge Graph — Coursera Notes › Organizations › Amazon › Amazon Web Services
Amazon Bedrock
service · part of Amazon Web Services
Fully managed service offering foundation models from multiple providers through a single API, with fine-tuning, agents, guardrails and knowledge bases.
Inside Amazon Bedrock (14)
- AI agent — A generative AI-powered system in Amazon Bedrock that can plan, reason, and interact to carry out complex tasks.
- Amazon Nova — Multimodal Amazon models for document/video analysis, content generation and agent building.
- Amazon Titan — Amazon's own family of foundation models available in Bedrock.
- Bedrock Agents — Generative-AI systems that plan, reason and call APIs/Lambda to complete multi-step tasks.
- Bedrock Flows — Bedrock Flows is a feature in Amazon Bedrock that enables the creation of structured, deterministic pipelines for orchestrating AI workflows.
- Custom Models & Fine-tuning — Adapting a foundation model to your domain using your own data.
- direct Bedrock API integration — Direct Bedrock API integration allows users to invoke AI models via the Amazon Bedrock API without writing code, typically through a visual interface or SDK.
- Guardrails — Safety filters that keep model output on-topic and free of unsafe or biased content.
- Knowledge Bases — Managed retrieval stores that let agents ground answers in your documents (RAG).
- PartyRock — A no-code playground built on Amazon Bedrock for quickly assembling generative AI applications.
- Request batching — Combining multiple inputs into a single API call to reduce per-call billing.
- Reserved capacity — A cost optimization option in Amazon Bedrock that allows customers to reserve capacity for a discounted rate.
- Response caching — Technique to store and reuse results of identical API calls to eliminate duplicate requests.
- Token — The billing unit for Amazon Bedrock, representing units of text that vary by model.
Connections
- Uses Foundation Models
- Uses AWS IAM
- Alternative to Amazon Q
- Alternative to Generative AI
- Used for Foundation Models
- Used for Large Language Models
- Uses AWS Lambda
- Uses Amazon Nova
- Offered by Amazon Web Services
- Used for Generative AI
- Uses Reserved capacity
- Used for Prompt injection
- Uses REST API
- Uses Response caching
- Uses Request batching
- Uses Lean prompts
- Used for Custom Models & Fine-tuning
- Related to AI agent
- Related to Data Wrangler
- Related to Foundation Models
- Related to Bedrock Agents
- Related to Guardrails
- Related to Knowledge Bases
- Related to PartyRock
Also known as: Amazon Bedrock, Bedrock
This is the text view of an interactive 3D knowledge graph — open this page with JavaScript enabled to explore it visually.