Coding & app building

Cloudflare Clef

By developers.cloudflare.com

Cloudflare Clef is a strong fit for typed classification and routing in software, with a profile optimized for advanced users who value medium ease-of-use and medium output quality.

Best for: Typed classification and routing in software

What it is

Cloudflare's Clef and Clef-flash decision-model family for typed choices, scores and probabilities from text or images through Workers AI or released weights; it does not generate free-form prose.

In Choosely terms, this sits in the coding & app building lane and is commonly selected for typed classification and routing in software and image-aware decisions with bounded answers.

Pricing

Workers AI: Clef $0.240 and Clef-flash $0.090 per million input tokens. Both share the Workers AI allowance of 10,000 neurons/day; usage above that requires Workers Paid at $0.011 per 1,000 neurons. Local weights have no Cloudflare inference charge, but hardware and operation cost extra.

Basis: Usage BasedConfidence: VerifiedLast checked: October 2026

Why people pick it vs where it falls short

Why people pick it

  • Hosted Clef and Clef-flash share a typed decision API
  • Supports text, structured state and up to four embedded images on Workers AI
  • Apache-2.0 weights and inference code are available for local evaluation

Where it falls short

  • Local deployment requires substantial model hardware and engineering
  • Published quality and latency comparisons are Cloudflare's own claims
  • Self-serve fine-tuning is planned; current customer tuning is hands-on

When it is a strong fit

A strong match when your main priority is typed classification and routing in software and you need an advanced-friendly starting point.

Useful when your team values medium ease of use and fast execution over heavier setup.

Best when medium quality matters, but you still want a practical workflow rather than a complex implementation track.

How it compares in Choosely terms

  • Speed profile: Fast. This is best when you want momentum from prompt to usable output without heavy process overhead.
  • Ease profile: Medium for Advanced users. You can move quickly even if this is not your full-time specialty.
  • Control profile: High. Expect practical customization, but not an infinite-control architecture.
  • Pricing signal: Usage-based. Good for teams balancing capability with cost sensitivity.
Tradeoff: Local deployment requires substantial model hardware and engineering.

Best-fit use cases

Practical ways Cloudflare Clef fits the current Choosely catalog profile.

Classify And Route Support Tickets With Typed Probabilities

Use Cloudflare Clef for classify and route support tickets with typed probabilities when you want fast execution, medium ease of use, and medium output quality.

Score Agent Decisions With Bounded Answers

Use Cloudflare Clef for score agent decisions with bounded answers when you want fast execution, medium ease of use, and medium output quality.

Classify Images For An Application

Use Cloudflare Clef for classify images for an application when you want fast execution, medium ease of use, and medium output quality.

Self Host An Open Weight Classifier

Use Cloudflare Clef for self-host an open-weight classifier when you want fast execution, medium ease of use, and medium output quality.

Alternatives

Jev by TypeSafe AI

TypeSafe AI's hosted early-access System One model for low-latency typed choices, scores and calibrated probabilities inside software; it does not generate open-ended prose.

Choose Jev by TypeSafe AI when your primary need is high-volume classification and routing inside applications.

ChatGPT

General-purpose conversational assistant for drafting, research, file work and coding help. Eligible users can also create a dot: an ongoing agent inside ChatGPT with a separate cloud computer and permissioned app access. Selected U.S. law firms can separately request early Trusted Access to GPT-6 Astra Law.

Choose ChatGPT if output quality consistency matters more than raw speed.

Google Gemini

Google's general conversational assistant for drafting, research support, multimodal work, planning, and broad productivity tasks. Gemini 4 Argon is a restricted family model currently available to trusted cyber defenders through the Fairwind Program, with wider access announced for later.

Choose Google Gemini when your primary need is general productivity.

Next step

Try Clef-flash and Clef on labeled decisions, compare them with Jev, then measure calibration, errors, image limits and cost before using decisions to trigger actions.

Related reads

FAQ

What is Cloudflare Clef best for?

Cloudflare Clef is best for typed classification and routing in software, image-aware decisions with bounded answers, teams evaluating open-weight decision models.

Is Cloudflare Clef beginner-friendly?

This catalog profile lists Cloudflare Clef at advanced skill level with medium ease of use.

What should I watch out for before choosing Cloudflare Clef?

Local deployment requires substantial model hardware and engineering