Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Cloudflare Clef is a structured decision model: your application supplies state and typed questions, and Clef returns probabilities for the answers your schema allows. Your code—not the model—then decides what those estimates mean, including when to act, fall back, or ask a person to review a case.
What Cloudflare Clef does
Clef is a cloud-hosted, 27-billion-parameter multimodal decision model available through Cloudflare Workers AI. Instead of producing open-ended text, it evaluates an input state—such as text, JSON, images, or video—and returns a probability for each allowed answer to each question. Cloudflare lists a 65,536-token context window and vision support on its Clef model page.
This constrained output makes Clef a possible fit when an application needs a bounded estimate: route a support case, classify an alert, score severity against a rubric, or check a guardrail. A probability is an estimate, not a guarantee that an answer is correct. Your application still needs to define the action policy around it.
How Clef’s question types work
A request can contain 1–64 questions. Each question uses one of three types, documented by Cloudflare:
#1 Best Overall
- 【DeskPi RackMate T1】It's made of aluminum alloy and acrylic frame mini chassis which you can setup your own cluster or home assistant server. For 10 inch 4U Server Cabinet (DeskPi RackMate T0), please refer to ASIN B0DPGZPTPP. For 10 inch 12U Server Cabinet (DeskPi RackMate T2), please refer to ASIN B0DT2XM22G.
- 【10-inch width】The cabinet has a width of 10 inches, which is a relatively small size that saves space while accommodating sufficient equipment. With dimensions of 11x7.8x16 inches, it is suitable for small offices, home environments, and large enterprises looking to save space.
- 【Open Design】The cabinet adopts an open design, allowing easy access to all devices inside. This design facilitates equipment installation and maintenance, aids in device cooling, and maintains optimal working conditions.
- 【8U Standard】The cabinet has a height of 8U, which is a standard unit size. With 1U equaling 1.75 inches, 8U implies a height of 14 inches.
- 【Translucent Design】Both sides are made of translucent acrylic, providing dust resistance and reduced weight. This design allows direct observation of the cabinet's interior, and users can add ambient lights for decoration.
| Type | What you define | What Clef returns |
|---|---|---|
noul |
A yes-or-no question | The probability of “yes” |
choice |
A set of permitted options | Probabilities for the options and a confidence value |
score |
An ordered scoring rubric | A probability-weighted score |
For example, a support workflow might pass an incident description as state, ask whether the incident is urgent, select among billing, technical, and sales teams, and rate severity on an ordered scale. The model estimates the answers; the application can then route the ticket or send it for review according to its own rules.
How to call Clef from a Worker
Cloudflare’s Workers AI binding example calls the model as @cf/cloudflare/clef. The Workers binding form is:
env.AI.run("@cf/cloudflare/clef", { /* state and typed questions */ })Use the schema and request shape in Cloudflare’s current model documentation for the state and questions you want to submit. The model page also documents an account REST endpoint using /ai/run; the exact endpoint and authentication details depend on that API path.
Images and other inputs
The model page lists text, JSON, images, and video as supported state formats. For embedded image input, the extension accepts up to four PNG, JPEG, or WebP images, subject to documented per-image and total-request limits. Remote image URLs are not accepted by this extension, so send supported image data rather than assuming a URL will be fetched. Check the live documentation for the current limits before building around them.
Set policy around probabilities
A useful decision flow separates the model’s estimate from the action that follows. Establish the policy in application code before treating an answer as an automated decision.
Rank #2
- COMPATIBILITY: Specially designed to mount Ubiquiti UniFi Cloud Gateway Fiber models UCG-Fiber and UXG-Fiber (30W) securely in place
- RACK SPECIFICATIONS: Standard 1U height rack mount bracket engineered for 10-inch rack installations, offering efficient space utilization
- MOUNTING SOLUTION: Provides stable and secure placement for your UniFi Cloud Gateway Fiber device in server room or network cabinet setups
- PACKAGE CONTENTS: Includes one (1) 1U 10-inch rack mount bracket specifically designed for UniFi Fiber Gateway installations
- INSTALLATION: Purpose-built bracket ensures proper device positioning and reliable mounting in standard 10-inch rack environments
- Define the allowed outcomes. Make choice options mutually understandable and make score rubrics ordered and operationally meaningful.
- Choose action thresholds. Specify which probability ranges permit an automatic action, which trigger a fallback, and which require human review. There is no universal threshold supplied by Clef.
- Handle uncertainty and missing results. Decide how the application behaves when the estimate is too close to call, the request fails, or an answer does not meet the schema expectations.
- Evaluate on representative cases. Use labeled examples from the task and inspect both overall quality and consequential error types before enabling automation.
- Monitor outcomes. Track model responses and downstream results so that drift, unexpected routing, or problematic edge cases can be caught and the policy revised.
For high-impact decisions, a structured probability should support a controlled workflow rather than silently replace accountable review.
Clef or Clef-flash?
Cloudflare’s October 1, 2026 launch announcement says both models are available on Workers AI. Its selection guidance positions the 27B Clef for higher-precision decisions and the 9B Clef-flash for latency-critical, hot-path decisions; both are listed with 64K-token contexts. The model page specifies Clef’s context as 65,536 tokens, the same nominal scale. These are vendor descriptions, so validate the trade-off on your own tasks.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware match| Model | Size | Cloudflare’s stated fit | Context listed in launch announcement |
|---|---|---|---|
| Clef | 27B | Highest-precision decisions | 64K tokens |
| Clef-flash | 9B | Latency-critical, hot-path decisions | 64K tokens |
Cloudflare’s announcement reports median latency across 43 benchmark runs of 209.3 ms for Clef and 38.8 ms for Clef-flash; the corresponding 95th-percentile figures were 238.6 ms and 122.4 ms. Those figures describe Cloudflare’s benchmark runs, not a latency guarantee for your deployment.
What Cloudflare’s benchmark results show—and do not show
Cloudflare’s launch post reports results across decision benchmarks and says a Clef model scored highest in 7 of 10. The following figures are vendor-reported; metric values should be read within each benchmark, not compared across different benchmarks.
| Benchmark and metric | Clef | Clef-flash | Jev |
|---|---|---|---|
| BFCL, case exact | 98.47 | 98.76 | 95.75 |
| BANKING77, macro-F1 | 94.20 | 90.93 | 79.74 |
| CLINC150+OOS, macro-F1 | 97.43 | 66.77 | 89.27 |
| Home appliances, case exact | 82.95 | 97.73 | 52.27 |
Cloudflare also reports 524.1 ms median and 536.0 ms 95th-percentile latency for Jev across the same announcement’s 43 benchmark runs. In a separate browser-based threat-classification workflow paired with Browser Run, Cloudflare reports 2.2 seconds for its workflow versus 4.7 seconds for gpt-oss-120b. Neither result establishes performance for other workloads or setups. See the full Cloudflare Workers AI announcement, dated October 1, 2026 for methodology and its qualifications.
Rank #3
- 【Powerful Load-bearing】12U Network Rack Open Frame is constructed from durable cold rolled steel; Rack shelf supports enhance stability, wall-mounted capacity of 130lbs, the ground-mounted up to 260lbs
- 【Considerate Designs】Open-frame layout, including a top panel adding space, anti-slip shelf stops fixing devices and compatible racks for stack and expansion to meet requirements of home server rack
- 【Complete Accessories】A 12U open frame server rack, two ventilated shelves, four shelf stops, four velcro straps and a set of equipment mounting screws
- 【Versatile Application】Ideal for space-efficient multi-device setups in warehouses, retail, classrooms, offices and more; Excellent choices as AV Rack/IT Rack
- 【Effortless Setup】 Network Rack includes hardware, a comprehensive manual, mounting hole drilling template and an online assembly video to simplify setup
For a deployment decision, test representative labeled cases and compare task quality, tail latency, input and context needs, hosted versus self-managed deployment requirements, cost, and integration compatibility. Vendor benchmark results are a useful starting point, not independent validation of your application’s outcomes.
Availability, cost, and migration considerations
Cloudflare identifies Workers AI as the hosted route for Clef and lists a price of $0.24 per million input tokens on the model page. Pricing and availability can change; check the live page for current terms before estimating operating cost.
Cloudflare says the Clef weights are published on Hugging Face under Apache 2.0. That statement does not by itself establish the deployment work or operating requirements for a self-managed setup; assess those separately against your infrastructure needs.
Cloudflare says Clef follows the System One API used by Typesafe’s Jev and describes an existing Jev integration as replaceable by changing the endpoint and model. Treat that as a migration path to investigate, not proof that every integration will behave equivalently. Test the actual schemas, outputs, thresholds, and operational assumptions before switching.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
