Best LLM Inference Providers

Vendor profileChecked Nov 8, 2026

Amazon Bedrock

Amazon Bedrock, a service of Amazon Web Services (AWS) · aws.amazon.com · Last checked:

  • Cloud provider platform
  • Serverless per-token API
  • Provisioned throughput
  • Bring-your-own-model (Custom Model Import)

Company status

Active. Amazon Bedrock is a service and business line of Amazon Web Services (Amazon.com, Inc.).

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Capabilities

  • Proprietary and open-weight models from many model providers (for example Anthropic, Meta, Mistral AI, OpenAI, Qwen, DeepSeek) billed per token by model and Region
  • OpenAI-compatible Chat Completions API on the bedrock-runtime and bedrock-mantle endpoints
  • Custom Model Import for customized open-weight models
  • Batch inference for select models
  • Per-Region data retention modes, including zero data retention (mode "none") for models that allow it; settings apply per AWS Region

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Who it tends to fit

  • Teams already on AWS that want IAM, VPC/PrivateLink, and AWS billing around model access
  • Buyers who need proprietary frontier models and open-weight models behind one cloud account

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Pricing

Published per-token pricing on https://aws.amazon.com/bedrock/pricing/ (by model and Region). AWS states batch inference for select models is priced 50% lower than on-demand.

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Published certifications

  • SOC (in scope)
  • ISO (in scope)
  • CSA STAR Level 2
  • HIPAA eligible
  • GDPR (AWS states customers can use Bedrock in compliance with the GDPR)
  • FedRAMP High (AWS GovCloud US-West)

AWS's Bedrock security page says Bedrock is in scope for ISO and SOC without naming report types; ask AWS Artifact for the specific SOC 2 / ISO 27001 reports.

Listed as stated by the vendor. Scope can differ by product or subsidiary; request current reports.

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Limitations to check

  • Retention depends on the model: some models require AWS retention or AWS human review (aws_review mode, up to 30 days) and are unavailable under zero-retention mode unless approved
  • Data retention settings are per Region and do not propagate to other Regions
  • Model availability and prices vary by AWS Region

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Sources

  1. Pricing page lists on-demand per-token prices by provider and Region; batch inference 50% lower than on-demand for select models; Custom Model Import pricing sectionaws.amazon.com/bedrock/pricing/Checked 2026-11-08
  2. Bedrock in scope for ISO, SOC, CSA STAR Level 2; HIPAA eligible; usable in compliance with GDPR; FedRAMP High in AWS GovCloud (US-West)aws.amazon.com/bedrock/security-compliance/Checked 2026-11-08
  3. Data retention modes (none, default, aws_review); per-Region configuration; aws_review retention up to 30 days; ZDR access for retention-requiring models by approvaldocs.aws.amazon.com/bedrock/latest/userguide/data-retention.htmlChecked 2026-11-08
  4. OpenAI Chat Completions API supported with Bedrock modelsdocs.aws.amazon.com/bedrock/latest/userguide/inference-chat-completions.htmlChecked 2026-11-08
  5. Custom Model Import imports customized open-source models into Bedrockdocs.aws.amazon.com/bedrock/latest/userguide/model-customization-import-model.htmlChecked 2026-11-08

Report a correction: [email protected]. Facts are re-checked at least every 90 days.