Post-Cutoff

CloudflareClef familyCurrentLanguage modelReleased 1 Oct 2026

Clef-flash

Fast open-weights decision model from Cloudflare. Same Jev-compatible API as cloudflare-clef. Hugging Face repo name taken from Cloudflare’s blog.

Model id@cf/cloudflare/clef-flash

Price$0.038 in per 1M tokens; output price not published

Checked10 October 2026

No published training cutoff

At least1,298events since its release on 1 October 2026

57 historic, 291 major. 49 days of news.

Cloudflare doesn’t publish a training cutoff, so we count from its release date. The real gap is larger.

Five things it has never heard of

How to call it

2 ways to use Clef-flash, with the model id each provider expects.

Providers, model ids, endpoints and docs for Clef-flash
ProviderModel idEndpointDocs
Cloudflare Workers AI @cf/cloudflare/clef-flash developers.cloudflare.com
Hugging Face huggingface.co

Pricing

Prices for Clef-flash
Input$0.038 per 1M tokens

As published: “per 1M input tokens (hosted, from Oct 9, 2026; was $0.09)”.

Source: blog.cloudflare.com, checked 10 October 2026.

Specs

Input
Text, images and video
Output
Text
Context window
24,000 tokens
Max output
Not published
Open weights
Yes
Licence
Apache-2.0
Training cutoff
Not published
Released
1 October 2026
Status
Current
Type
Language model

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Low-latency decision model

    Qwen3.5-9B-based option scorer; Cloudflare reports 38.8 ms median and 122.4 ms p95 latency on its 43-benchmark set and top scores on BFCL (98.76) and API-Bank (93.11).

    Source: blog.cloudflare.com

Notes

Smaller, faster sibling of Clef (Qwen3.5-9B backbone). Self-hosting needs ~41 GB VRAM at 64k context (The Register). Oct 9, 2026: hosted price cut 58% to $0.038/M input tokens (from $0.09), below Jev, and hosted context cut from 64k to 24k (Cloudflare: only 0.24% of requests exceeded 24k); Hugging Face weights unchanged. Same release made it ~1.7-2x faster (152 ms vs 262 ms median at ~800 tokens).