Amazon Bedrock vs Cloudflare Workers AI

Bedrock brings Claude, GPT and open models under AWS security controls. Cloudflare Workers AI serves open models only, on its own network, with a simpler bill.

By The Subconscious Team · Updated September 30, 2026

Amazon Bedrock vs Cloudflare Workers AI: key differences

Bedrock is a control plane for 100+ models from 18+ providers, including Claude, OpenAI models up to GPT-6 Astra, Amazon Nova and DeepSeek. Every call inherits IAM, PrivateLink, KMS encryption and CloudTrail logging. Cloudflare serves 50+ open models on GPUs inside its own network, with no closed models in the catalog. Pricing philosophies differ. Third-party analyses put most Bedrock models about 20 to 35% above direct API prices, with Claude at parity, though batch cuts 50% and provisioned throughput cuts 20 to 40%. Cloudflare publishes per-token equivalents that track first-party lists on several models: DeepSeek V4 Pro at $1.32 in and $3.96 out matches DeepSeek's own peak rate, and GLM 5.3 matches Z.ai's $1.40 in and $4.40 out.

Customization and agents favor Bedrock. It offers fine-tuning, Custom Model Import for open checkpoints, Knowledge Bases for RAG, Guardrails for PII redaction, and AgentCore with memory, an MCP gateway, sandboxed tools and Cedar policies. Cloudflare limits customization to bring-your-own LoRA on smaller models in beta, and its large models need the Workers Paid plan. Cloudflare wins on simplicity: an OpenAI-compatible endpoint, 10,000 free Neurons a day, and AI Gateway for caching and fallbacks without idle add-on charges. Bedrock model availability also varies by region. Regulated AWS shops should stay on Bedrock. Teams building on Workers that only need open models will find Cloudflare cheaper to start and simpler to run.

What Amazon Bedrock and Cloudflare Workers AI do

Amazon Bedrock

Amazon Bedrock is AWS's managed model service and has become the default AI control plane for many enterprises. One API reaches 100+ models from 18+ providers, including Anthropic's Claude family, Meta, Mistral, DeepSeek, Amazon's own Nova models, and, since an April 2026 partnership expansion, OpenAI models up to GPT-6 Astra. Switching models is usually just a new model ID. Every call inherits IAM, PrivateLink, KMS encryption and CloudTrail logging, and provider models never train on customer data.

Example models: Claude Opus, GPT-6 Astra

Full Amazon Bedrock profile

Cloudflare Workers AI

Workers AI is the serverless GPU inference service of Cloudflare, which was founded in 2009. It launched in September 2023 and reached general availability in April 2024. Models run on GPUs inside Cloudflare's own network and are called from a Worker through an AI binding or over REST, including OpenAI-compatible Chat Completions and Embeddings endpoints plus a Responses endpoint for gpt-oss. The catalog lists 50+ open models. Since Kimi K2.5 arrived in March 2026 it has carried frontier-scale LLMs: Kimi K2.6 and K2.7 Code, GLM 5.2 and 5.3, DeepSeek V4 Pro and Flash, gpt-oss 120B and 20B, Qwen 3.8 27B and Llama 4 Scout. DeepSeek V4, added August 14, 2026, was the first to offer the full 1,048,576 token context.

Example models: DeepSeek V4 Pro, GLM 5.3

Full Cloudflare Workers AI profile

Should you choose Amazon Bedrock or Cloudflare Workers AI?

Amazon Bedrock

Choose Amazon Bedrock for

Regulated enterprises already inside AWS
Claude and GPT alongside open models on one API
Governed agents on AgentCore with audit trails

Cloudflare Workers AI

Choose Cloudflare Workers AI for

Open models at first-party list prices
Teams without an AWS security posture to inherit
Lightweight inference inside Workers apps

Amazon Bedrock vs Cloudflare Workers AI at a glance

Attribute	Amazon Bedrock	Cloudflare Workers AI
Model access	Closed and open, 100+ models	Open weights
Flagship models	Claude, GPT-6 Astra, Nova, DeepSeek	DeepSeek V4 Pro, GLM 5.3, Kimi K2.7 Code, gpt-oss 120B
Speed	Latency-optimized option on some models	Unknown
Price	~20–35% above direct; Claude at parity	$0.011 per 1K Neurons; 10K free daily
Customization	Fine-tuning, Custom Model Import	BYO LoRA on small models (beta)
Deployment	Managed on AWS, AgentCore	Serverless on Cloudflare network
Long context	Varies by model	1M on DeepSeek V4; 262K on Kimi

Frequently asked questions

What is the difference between Amazon Bedrock and Cloudflare Workers AI?

Bedrock brings Claude, GPT and open models under AWS security controls. Cloudflare Workers AI serves open models only, on its own network, with a simpler bill.

When should I choose Amazon Bedrock over Cloudflare Workers AI?

Regulated enterprises already inside AWS; Claude and GPT alongside open models on one API; Governed agents on AgentCore with audit trails.

When should I choose Cloudflare Workers AI over Amazon Bedrock?

Open models at first-party list prices; Teams without an AWS security posture to inherit; Lightweight inference inside Workers apps.

Is Amazon Bedrock or Cloudflare Workers AI cheaper?

Amazon Bedrock: ~20–35% above direct; Claude at parity. Cloudflare Workers AI: $0.011 per 1K Neurons; 10K free daily. The cheaper choice depends on the model and workload.

Which has more context, Amazon Bedrock or Cloudflare Workers AI?

Amazon Bedrock: Varies by model. Cloudflare Workers AI: 1M on DeepSeek V4; 262K on Kimi.

Related comparisons

Running long-horizon agents?

If your agents run past 200K tokens, compare both against Subconscious. Our inference stack treats a long-horizon trace as the primary workload, so speed, cost, and accuracy hold up deep into the trace.

Subconscious vs Amazon Bedrock Subconscious vs Cloudflare Workers AI