macrostack

AI API price index · cloud platform · read 14 Nov 2026

Amazon Bedrock API pricing

25 prices as Amazon Bedrock publishes them — 24 text models, 1 embeddings — read on its own pricing page on 14 Nov 2026. Its current text models cost $0.14 to $15.00 per million output tokens.

Free tier No free token allowance is stated on the Bedrock pricing page. AWS's general Free Tier credit is an account-wide offer, not a Bedrock price.

What to know before you build on Amazon Bedrock

  • Bedrock is AWS's model marketplace: Amazon's own Nova models plus models from OpenAI, Meta, Moonshot, DeepSeek, Mistral, xAI, Qwen and others, billed on the AWS account.
  • The newer models are cheapest through Global cross-Region inference; keeping a call in one Region or in the US costs 10% more (GPT-6 Astra $10 and $50 per 1M tokens global, $11 and $55 in the US).
  • OpenAI's GPT-6 models on Bedrock cost the same as on OpenAI's own API at the global rate, and so does xAI's Grok 4.7.
  • Flex and Batch cost half of Standard where a model offers them; Priority costs more.
  • Prices differ by Region: the pricing page opens on Sydney, where gpt-oss-120b is $0.1545 in and $0.618 out against $0.15 and $0.60 in N. Virginia. The index uses N. Virginia.

Against the other 31

Across the index, 390 current text models run from $0.03 to $50.00 per million output tokens — 1667×. Amazon Bedrock’s cheapest, Nova Micro at $0.14, has 11 cheaper text models elsewhere; its dearest, Kimi K3, costs $15.00. A cheaper model is a different model, not the same one for less — compare capability before price.

The same model elsewhere

Open-weight models Amazon Bedrock sells that other hosts price too — the only like-for-like comparison in the index.

  • gpt-oss-120b — Novita AI $0.05 / $0.25 · Google Vertex AI (Agent Platform) $0.09 / $0.36 · SambaNova Cloud $0.22 / $0.59 · Amazon Bedrock $0.15 / $0.60 · Fireworks AI $0.15 / $0.60 · Groq $0.15 / $0.60 · Nebius Token Factory $0.15 / $0.60 · Together AI $0.15 / $0.60 · Cerebras $0.35 / $0.75 · Cloudflare Workers AI $0.35 / $0.75 — 3.0× apart
  • Llama 3.3 70B — Novita AI $0.135 / $0.40 · Azure AI Foundry Models $0.71 / $0.71 · Amazon Bedrock $0.72 / $0.72 · Google Vertex AI (Agent Platform) $0.72 / $0.72 · SambaNova Cloud $0.60 / $1.20 — 3.0× apart
  • gpt-oss-20b — Google Vertex AI (Agent Platform) $0.07 / $0.25 · Amazon Bedrock $0.07 / $0.30 · Cloudflare Workers AI $0.20 / $0.30 · Groq $0.075 / $0.30 — 1.2× apart
  • Kimi K3 — DeepInfra $2.85 / $14.25 · Amazon Bedrock $3.00 / $15.00 · Fireworks AI $3.00 / $15.00 · Moonshot AI (Kimi) $3.00 / $15.00 · Nebius Token Factory $3.00 / $15.00 · Novita AI $3.00 / $15.00 · Perplexity $3.00 / $15.00 · Together AI $3.00 / $15.00 — 1.1× apart

Input / output per 1M tokens, standard tier.

The same model at its maker

Amazon Bedrock resells models other companies make. Here is each one beside its maker's own price, both at the standard global rate and the standard prompt length.

ModelAmazon BedrockAt its makerDifference
grok-4.7$2.00 / $6.00$2.00 / $6.00 xAIThe same price
gpt-6-astra$10.00 / $50.00$10.00 / $50.00 OpenAIThe same price
gpt-6.1-sol$2.00 / $10.00$2.00 / $10.00 OpenAIThe same price
gpt-6.1-terra$1.00 / $5.00$1.00 / $5.00 OpenAIThe same price
gpt-6-sol$2.00 / $10.00$2.00 / $10.00 OpenAIThe same price
gpt-6-luna$0.10 / $0.50$0.10 / $0.50 OpenAIThe same price

Input / output per 1M tokens (embeddings: input only). 6 of 6 cost the same here as at their maker. The index ranks each of these models once, at its maker’s price.

Text models

ModelInputCached inputOutputNote
Nova Micro$0.035$0.00875$0.14On-demand, in-Region (us-east-1)
Nova Lite$0.06$0.015$0.24On-demand, in-Region (us-east-1)
Nova Pro$0.80$0.20$3.20On-demand, in-Region (us-east-1)
Nova Premier$2.50$0.625$12.50On-demand, in-Region (us-east-1)
Nova 2.0 Lite$0.30$0.075$2.50On-demand, Global cross-Region inference
Nova 2.0 Litein-region$0.33$0.0825$2.75On-demand, in-Region (us-east-1) · Not ranked: in-Region price (10% above the global rate); the index ranks the global price.
Nova 2.0 Pro$1.25—$10.00On-demand, Global cross-Region inference
Nova 2.0 Proin-region$1.375—$11.00On-demand, in-Region (us-east-1) · Not ranked: in-Region price (10% above the global rate); the index ranks the global price.
gpt-oss-120b$0.15—$0.60On-demand, in-Region (us-east-1)
gpt-oss-20b$0.07—$0.30On-demand, in-Region (us-east-1)
Llama 3.3 70B$0.72—$0.72On-demand, in-Region (us-east-1)
Llama 4 Maverick 17B$0.24—$0.97On-demand, in-Region (us-east-1)
Kimi K3$3.00$0.30$15.00On-demand, Global cross-Region inference
Kimi K3in-region$3.30$0.33$16.50On-demand, US cross-Region inference · Not ranked: in-Region price (10% above the global rate); the index ranks the global price.
DeepSeek v3.2$0.62—$1.85On-demand, in-Region (us-east-1)
GLM 5$1.00—$3.20On-demand, in-Region (us-east-1)
Qwen3 235B A22B 2507$0.22—$0.88On-demand, in-Region (us-east-1)
Mistral Large 3$0.50—$1.50On-demand, in-Region (us-east-1)
grok-4.7$2.00$0.50$6.00On-demand, Global cross-Region inference · Not ranked: the same model xAI sells; ranked once, at xAI's own price.
gpt-6-astra$10.00$1.00$50.00Short context, Global cross-Region inference (OpenAI GPT-6 Astra, Amazon Bedrock Edition) · Not ranked: the same model OpenAI sells; ranked once, at OpenAI's own price.
gpt-6.1-sol$2.00$0.10$10.00Short context, Global cross-Region inference (OpenAI GPT-6.1 Sol, Amazon Bedrock Edition) · Not ranked: the same model OpenAI sells; ranked once, at OpenAI's own price.
gpt-6.1-terra$1.00$0.05$5.00Short context, Global cross-Region inference (OpenAI GPT-6.1 Terra, Amazon Bedrock Edition) · Not ranked: the same model OpenAI sells; ranked once, at OpenAI's own price.
gpt-6-sol$2.00$0.20$10.00Short context, Global cross-Region inference (OpenAI GPT-6 Sol, Amazon Bedrock Edition) · Not ranked: the same model OpenAI sells; ranked once, at OpenAI's own price.
gpt-6-luna$0.10$0.01$0.50Short context, Global cross-Region inference (OpenAI GPT-6 Luna, Amazon Bedrock Edition) · Not ranked: the same model OpenAI sells; ranked once, at OpenAI's own price.

Embeddings

ModelInputCached inputNote
Titan Text Embeddings V2$0.02—On-demand, us-east-1

16 of these 25 prices are ranked in the index; the rest say why in their note. Text and embeddings per 1M tokens unless the row says otherwise.

Discounts and surcharges, as published

  • Service tiers: "Standard, Priority, Flex, and Reserved" (pricing page). Flex and Batch are half of Standard where offered: gpt-oss-120b in Sydney is $0.1545 in and $0.618 out on Standard, $0.0773 and $0.309 on Flex and Batch (read on the pricing page 14 Nov 2026). Priority costs more.
  • Batch: "select foundation models … for batch inference at a 50% lower price compared to on-demand" (pricing page).
  • Global cross-Region inference is the cheaper way to call the newer models: in-Region or US-only calls cost 10% more (Nova 2.0 Lite $0.33 in and $2.75 out per 1M tokens in-Region against $0.30 and $2.50 global; GPT-6 Astra $11 and $55 against $10 and $50).
  • Provisioned Throughput is sold per model unit per hour (Nova Pro $60.50 an hour with no commitment, $30.25 on a six-month term); it is a reservation, not a per-token price.

Where these numbers come from

Read on aws.amazon.com/bedrock/pricing on 14 Nov 2026, checked against 2 more of Amazon Bedrock’s own pages. Macrostack measures nothing and is paid by no provider on this index. Prices change; the date is the claim.

The other providers in the index

Questions

How much does the Amazon Bedrock API cost?
On 14 Nov 2026, Amazon Bedrock's current text models cost $0.14 to $15.00 per million output tokens at the standard tier (Nova Micro to Kimi K3). The full list, with input and cached-input prices, is on this page.
Is Amazon Bedrock cheaper than buying from the model maker directly?
No, and it is not dearer either: on 14 Nov 2026 grok-4.7 cost $2.00 in and $6.00 out per million tokens on Amazon Bedrock, against $2.00 and $6.00 at xAI; gpt-6-astra cost $10.00 in and $50.00 out per million tokens on Amazon Bedrock, against $10.00 and $50.00 at OpenAI; gpt-6.1-sol cost $2.00 in and $10.00 out per million tokens on Amazon Bedrock, against $2.00 and $10.00 at OpenAI. The reasons to choose one over the other are billing, data residency, compliance and the rest of the cloud you already use, not the token price.
Does Amazon Bedrock have a free tier?
What Amazon Bedrock's own pages say: No free token allowance is stated on the Bedrock pricing page. AWS's general Free Tier credit is an account-wide offer, not a Bedrock price.
Does Amazon Bedrock offer discounts?
Service tiers: "Standard, Priority, Flex, and Reserved" (pricing page). Flex and Batch are half of Standard where offered: gpt-oss-120b in Sydney is $0.1545 in and $0.618 out on Standard, $0.0773 and $0.309 on Flex and Batch (read on the pricing page 14 Nov 2026). Priority costs more. Batch: "select foundation models … for batch inference at a 50% lower price compared to on-demand" (pricing page). Global cross-Region inference is the cheaper way to call the newer models: in-Region or US-only calls cost 10% more (Nova 2.0 Lite $0.33 in and $2.75 out per 1M tokens in-Region against $0.30 and $2.50 global; GPT-6 Astra $11 and $55 against $10 and $50). Provisioned Throughput is sold per model unit per hour (Nova Pro $60.50 an hour with no commitment, $30.25 on a six-month term); it is a reservation, not a per-token price.
The Macrostack brief

New swaps, worth your inbox.

A short, occasional email when we add a high-intent alternative or ship a new head-to-head. No spam, no selling your address — unsubscribe in one click.