ModelPriceWatch.com
Last scan 2026-10-13 Models tracked 286 Providers 39 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

Cloudflare

Commercial provider · 3 models tracked

The connectivity-cloud company, which runs Workers AI, a serverless inference platform on GPUs across its global network that hosts dozens of open models billed in "neurons". Only the models Cloudflare trained itself are tracked here: the Clef family of decision models (Clef, Clef-omni and Clef-flash), launched October 2026, which return a probability for each allowed answer to typed questions rather than generating text, are open-sourced under Apache 2.0, and bill input tokens only with no output charge.

Cloudflare pricing at a glanceOctober 2026 · $ per 1M tokens

Cloudflare API pricing (October 2026): 3 current models range from $0.038 to $0.240 per 1M input tokens. The cheapest paid model is Clef-flash at $0.038/1M input; the priciest is Clef at $0.240/1M input. Every price links to Cloudflare's official pricing page and refreshes twice daily.

Models3
Input range$0.038–$0.240
Cached-input tiers0

Today's Cloudflare prices

3 models · cheapest blended first

Sorted by blended cost (cheapest first). Prices per 1M tokens, October 2026 — every price links to Cloudflare's official pricing page.

Current Cloudflare model prices per 1M tokens, sorted by blended cost
Model Blended* Input Output Cached in Relative cost Context Status
Clef-flash
$0.028
$0.038 $0 —
25K Current
Clef-omni
$0.112
$0.150 $0 —
64K Current
Clef
$0.180
$0.240 $0 —
66K Current
* Blended = (3×input + 1×output) ÷ 4 $/1M tokens · cheap · mid · expensive MODELPRICEWATCH.COM · 2026-10-13

Quick stats

Models tracked
3
Type
commercial
Founded
—
Cheapest model
Clef-flash
Cheapest blended
$0.028/M

Open-weight models

3 open-weight models available from this provider.

Clef
$0.240/M in · 66K ctx
Clef-omni
$0.150/M in · 64K ctx
Clef-flash
$0.038/M in · 25K ctx
Run it yourself

Self-host Cloudflare's open-weight models

These models ship with open weights, so you can serve them yourself on rented GPUs instead of paying per-token API prices — often cheaper at scale.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.