Cloudflare Workers AI
Try Cloudflare Workers AI on us. Starter credit covers your first requests, then one key reaches all 28 Cloudflare Workers AI models, with the privacy posture shown on every route.
Cloudflare Workers AIcloudflare-workers-ai
These privacy labels describe Cloudflare Workers AI, the upstream model provider. Confidential requires all three: verified provider compute, end-to-end encryption and explicit zero data retention. Account and billing metadata are outside the content-retention claim.
| Provider | Cloudflare Workers AI |
|---|---|
| API operator country | US Legal home of the API operator, not an inference-location guarantee. |
| Headquarters | Not verified |
| Routing status | Active |
| Provider website | https://www.cloudflare.com/developer-platform/products/workers-ai/ |
| Models | 28 public models |
| Credits routes | 28 |
| Zero data retention | not claimed |
| Verified confidential inference | Not verified |
| Policy note | Workers AI documents no input/output retention after ordinary stateless inference and no model training. Stateful features, customer-enabled logging/caching, and safety, security or legal exceptions may retain content. No exception-free commitment is verified for this account; these routes remain outside strict ZDR. Policy source |
Inference locations
Provider API declarations are also available on each model's endpoint API. Direct chat and Responses replies separate advertised availability from serving location in usage.inference_location; the actual serving region remains unknown when not reported.
usage.provider_usage.gateway_region identifies the TrustedRouter gateway, not the provider's inference location. A value such as europe-west4 does not mean the model ran in Europe. usage.inference_location.serving_region remains null when the provider has not reported it. Older gateways may return the gateway-only field as provider_usage.region.
Where the upstream GPUs execute the model, not the provider's headquarters, API ingress, or data storage. ZDR and confidential inference do not by themselves restrict geography.
- Declared countries or regions
- Not verified
No inference-location declaration has been verified for this provider. - Can locations change?
- Not verified. Do not assume that the serving location is fixed.
- Provider region pinning
- Not verified.
- Pinning through TrustedRouter
- No verified provider-specific location pin documented here.
- Infrastructure declaration
- Inference location not verified; no country-specific residency commitment.
- Evidence
- Not yet reviewed.
Provider selection alone is not a region pin. TrustedRouter's US provider filter selects company jurisdiction, not GPU location. A regional gateway hostname does not pin a downstream GPU. For a binding residency requirement, confirm the model, region, failover scope and enforced controls before sending sensitive data.
Measured performance
31 samplesContinuously sampled across Cloudflare Workers AI's routed models: p50 TTFT, effective throughput, and success rate. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime. No prompt or output content stored.
| p50 TTFT | 2452 ms |
|---|---|
| Effective throughput | 72 tok/s n=8 |
| Uptime | 100.00% |
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| ibm-granite/granite-4.0-h-micro | 1441 ms | — | 100.00% | — | 1 |
| qwen/qwen2.5-coder-32b-instruct | 1510 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-v4.1-flash-1017 | 1517 ms | — | 100.00% | — | 1 |
| aisingapore/gemma-sea-lion-v4-27b-it | 1602 ms | — | 100.00% | — | 2 |
| qwen/qwq-32b | 1722 ms | — | 100.00% | — | 1 |
| mistralai/mistral-small-3.1-24b-instruct | 1844 ms | — | 100.00% | — | 2 |
| openai/gpt-oss-20b | 1997 ms | — | 100.00% | — | 2 |
| cloudflare/clairic | 2184 ms | — | 100.00% | — | 1 |
| z-ai/glm-5.2 | 2288 ms | 108 tok/s n=1 | 100.00% | — | 1 |
| meta-llama/llama-4-scout-17b-16e-instruct | 2387 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-r1-distill-qwen-32b | 2452 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k2.7-code | 2460 ms | 72 tok/s n=2 | 100.00% | — | 1 |
| meta-llama/llama-3.2-3b-instruct | 2621 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-v4-flash-0731 | 2670 ms | — | 100.00% | — | 1 |
| openai/gpt-oss-120b | 2694 ms | — | 100.00% | — | 2 |
| nvidia/nemotron-3-120b-a12b | 2723 ms | — | 100.00% | — | 1 |
| google/gemma-4-26b-a4b-it | 2805 ms | — | 100.00% | — | 1 |
| meta-llama/llama-3.3-70b-instruct-fp8-fast | 2851 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k3 | 2884 ms | 45 tok/s n=1 | 100.00% | — | 2 |
| moonshotai/kimi-k2.6 | 3125 ms | 43 tok/s n=2 | 100.00% | — | 1 |
| z-ai/glm-5.3 | 3703 ms | 93 tok/s n=2 | 100.00% | — | 1 |
| z-ai/glm-5.3-flash | 8571 ms | — | 100.00% | — | 1 |
Cloudflare Workers AI performance history · Full provider & model leaderboard.
Models served by Cloudflare Workers AI.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Input | Cached input | Output |
|---|---|---|---|---|---|
aisingapore/gemma-sea-lion-v4-27b-it@cf/aisingapore/gemma-sea-lion-v4-27b-it |
— | 128,000 | $0.370305/1M | Not published | $0.585525/1M |
cloudflare/clairicCloudflare Clairic |
— | 65,536 | $0.1055/1M | Not published | $0/1M |
deepseek/deepseek-r1-distill-qwen-32b@cf/deepseek-ai/deepseek-r1-distill-qwen-32b |
— | 80,000 | $0.524335/1M | Not published | $5.149455/1M |
deepseek/deepseek-v4-flash-0731DeepSeek: DeepSeek V4 Flash 0731 |
— | 1,048,576 | $0.4642/1M | $0.01477/1M | $1.3926/1M |
deepseek/deepseek-v4.1-flash-1017DeepSeek: DeepSeek V4.1 Flash 1017 |
IQ 128#19 | 1,048,576 | $0.3165/1M | $0.01/1M | $1.266/1M |
google/gemma-4-26b-a4b-itGoogle: Gemma 4 26B A4B |
IQ 99#131 | 262,144 | $0.1055/1M | $0.05275/1M | $0.3165/1M |
ibm-granite/granite-4.0-h-micro@cf/ibm-granite/granite-4.0-h-micro |
— | 131,000 | $0.017935/1M | Not published | $0.11816/1M |
meta-llama/llama-3.1-8b-instruct-fp8@cf/meta/llama-3.1-8b-instruct-fp8 |
— | 32,000 | $0.16036/1M | Not published | $0.302785/1M |
meta-llama/llama-3.2-11b-vision-instruct@cf/meta/llama-3.2-11b-vision-instruct |
— | 128,000 | $0.051168/1M | Not published | $0.71318/1M |
meta-llama/llama-3.2-1b-instruct@cf/meta/llama-3.2-1b-instruct |
— | 60,000 | $0.028485/1M | Not published | $0.212055/1M |
meta-llama/llama-3.2-3b-instructMeta: Llama 3.2 3B Instruct |
— | 131,072 | $0.0537/1M | Not published | $0.353425/1M |
meta-llama/llama-3.3-70b-instruct-fp8-fast@cf/meta/llama-3.3-70b-instruct-fp8-fast |
— | 24,000 | $0.309115/1M | Not published | $2.376915/1M |
meta-llama/llama-4-scout-17b-16e-instructLlama 4 Scout Instruct |
— | 131,072 | $0.28485/1M | Not published | $0.89675/1M |
mistralai/mistral-small-3.1-24b-instruct@cf/mistralai/mistral-small-3.1-24b-instruct |
— | 128,000 | $0.370305/1M | Not published | $0.585525/1M |
moonshotai/kimi-k2.6MoonshotAI: Kimi K2.6 |
IQ 114#62 | 262,144 | $1.00225/1M | $0.1688/1M | $4.22/1M |
moonshotai/kimi-k2.7-codeMoonshotAI: Kimi K2.7 Code |
IQ 111#77 | 262,144 | $1.00225/1M | $0.20045/1M | $4.22/1M |
moonshotai/kimi-k3MoonshotAI: Kimi K3 |
IQ 127#22 | 1,048,576 | $3.165/1M | $0.3165/1M | $15.825/1M |
nvidia/nemotron-3-120b-a12b@cf/nvidia/nemotron-3-120b-a12b |
— | 256,000 | $0.5275/1M | Not published | $1.5825/1M |
openai/gpt-oss-120bOpenAI: gpt-oss-120b |
IQ 98#136 | 131,072 | $0.36925/1M | Not published | $0.79125/1M |
openai/gpt-oss-20bOpenAI: gpt-oss-20b |
IQ 93#155 | 131,072 | $0.211/1M | Not published | $0.3165/1M |
qwen/qwen2.5-coder-32b-instruct@cf/qwen/qwen2.5-coder-32b-instruct |
— | 32,768 | $0.6963/1M | Not published | $1.055/1M |
qwen/qwen3-30b-a3b-fp8@cf/qwen/qwen3-30b-a3b-fp8 |
— | 32,768 | $0.0537/1M | Not published | $0.353425/1M |
qwen/qwen3.8-27bQwen: Qwen3.8 27B |
IQ 115#57 | 65,536 | $0.47475/1M | $0.05275/1M | $3.376/1M |
qwen/qwq-32b@cf/qwen/qwq-32b |
— | 24,000 | $0.6963/1M | Not published | $1.055/1M |
z-ai/glm-4.7-flashZ.ai: GLM 4.7 Flash |
— | 131,072 | $0.063828/1M | Not published | $0.422/1M |
z-ai/glm-5.2Z.ai: GLM 5.2 |
IQ 118#53 | 1,048,576 | $1.477/1M | $0.2743/1M | $4.642/1M |
z-ai/glm-5.3Z.ai: GLM 5.3 |
IQ 124#30 | 1,048,576 | $1.477/1M | $0.2743/1M | $4.642/1M |
z-ai/glm-5.3-flashZ.ai: GLM 5.3 Flash |
IQ 119#50 | 1,048,576 | $0.15825/1M | $0.03165/1M | $0.5275/1M |
Questions
Does Cloudflare Workers AI have zero data retention?
TrustedRouter does not currently mark Cloudflare Workers AI as provider-level zero data retention. Use trustedrouter/zdr or provider.min_privacy=zdr to select a different eligible route, and review the linked policy source for changes.
Is Cloudflare Workers AI end-to-end encrypted?
TrustedRouter does not currently mark Cloudflare Workers AI as end-to-end encrypted at the provider boundary. The TrustedRouter gateway is still attested, but the selected provider normally receives the request in order to run the model. Use trustedrouter/e2e for the stronger route requirement.
Which Cloudflare Workers AI models are available through TrustedRouter?
This page currently lists 28 public Cloudflare Workers AI models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.