Amazon Bedrock

Multi-vendor model access inside your existing AWS account

aws.amazon.com/bedrock docs.aws.amazon.com/bedrock

Overall in llm apis

4of 18 · score 73.1

Data coverage

93%

25 of 27 fields sourced

Best for

•Regulated workloads that must stay inside an existing AWS perimeter
•Australian data residency without self-hosting
•Teams that need a documented model lifecycle to satisfy an auditor

Strong on

+Effort control Yes 1/18
+Zero retention Yes 1/18
+AU region Yes 1/18
+Computer use Yes 1/18
+Notice period 180 days 1/7

Weak on

−OpenAI-compat API No 18/18
Better: OpenAI: Yes
−Open weights Partial 12/18
Better: Meta Llama: Yes
−$/M input $5/M tok 11/13
Better: Groq: $0.05/M tok
−$/M output $25/M tok 11/13
Better: Fireworks AI: $0.05/M tok

vs Google Gemini

•Ahead on Zero retention: Yes vs Partial
•Ahead on Computer use: Yes vs Partial
•Behind on OpenAI-compat API: No vs Yes
•Behind on Trains on your data: No vs Partial

How Amazon Bedrock scores among 18 llm apis25 of 27 fields known

Capability7

Flagship model

Multi-vendor, Claude Opus 4.8, Llama 4, Mistral, Nova Premier

docs.aws.amazon.com · 2026-06-24

Context window1st

1,000,000 tokens

1/14

· 2026-06-24

Max output

128,000 tokens

4/10

Leader: xAI · Unlimited

· 2026-06-24

Image input1st

Yes

1/18

docs.aws.amazon.com · 2026-06-24

Audio in/out1st

Partial

1/18

· 2026-01-15

Open weights

Partial

12/18

Leader: Meta Llama · Yes

· 2026-01-15

OpenAI-compat API

18/18

Leader: OpenAI · Yes

· 2026-06-24

Performance2

TTFT p50

—

Output tok/s

—

Pricing4

$/M input

$5/M tok

11/13

Leader: Groq · $0.05/M tok

platform.claude.com · 2026-11-07

$/M output

$25/M tok

11/13

Leader: Fireworks AI · $0.05/M tok

platform.claude.com · 2026-11-07

$/M cache read

$0.5/M tok

10/12

Leader: DeepSeek · $0.0028/M tok

platform.claude.com · 2026-11-07

Batch discount1st

50%

1/14

· 2026-01-15

Agentic6

Tool use1st

Yes

1/18

docs.aws.amazon.com · 2026-06-24

Schema output1st

Yes

1/18

docs.aws.amazon.com · 2026-06-24

Effort control1st

Yes

1/18

· 2026-06-24

Computer use1st

Yes

1/18

docs.aws.amazon.com · 2026-06-24

MCP support

6/18

Leader: Anthropic · Yes

· 2026-06-24

Cache TTL

5 min default, 1 h option; explicit breakpoints only

docs.aws.amazon.com · 2026-06-24

Governance & continuity6

Continuity policy

The most contractual answer in the category. Bedrock model IDs are versioned and move through a published Active to Legacy to end-of-life lifecycle with dated notices in the console and documentation, and legacy versions keep serving existing workloads after a successor lands. The cost of that stability is lag: new features reach Bedrock weeks to months after the first-party API, and some never do.

docs.aws.amazon.com · 2026-06-24

Notice period1st

180 days

1/7

docs.aws.amazon.com · 2026-11-07

Pinnable versions1st

Yes

1/18

docs.aws.amazon.com · 2026-06-24

Zero retention1st

Yes

1/18