Cloudflare model profile

Clef Flash decision model: measured support-routing profile

Clef Flash is the lower listed input-price Cloudflare starting point and had a 195 ms median in this run. On the 100 frozen cases it returned valid output every time, and its accuracy was 87% (79–92), with the 95% interval in parentheses. The intervals describe uncertainty in each model's observed rate. This report does not include a paired statistical comparison of the differences between models.

Datasetsupport-routing v1

Cases100 held-out

Model revisionclef-flash

Run dateOctober 9, 2026

Measured on support routing

Evidence noteDecision quality, validity, and latency were independently measured from the frozen run. Estimated cost per 1,000 tickets multiplies provider-reported token counts from this run by provider list prices last compared on October 9, 2026; it is not a billed or measured cost. Values in parentheses are the 95% interval at n=100.

Correct routes: 87/100
Accuracy: 87% (79–92)
Operational success: 87% (79–92)
Equal-label macro-F1: 86% (79–92)
Valid responses: 100/100
Median latency: 195 ms
P95 latency: 292 ms
Estimated cost per 1,000 tickets: ≈ $0.027

These numbers describe one balanced set of fictional support cases. They are not a general model ranking or a production service-level claim.

Clarification requests

Exploratory, n=20

Clef Flash sent 10 of 20 held-out needs_clarification cases back for clarification instead of routing them to a team: recall 50% (30–70), shown with its 95% Wilson interval. The denominator includes all 20 cases labeled needs_clarification, including attempts with no valid answer. This slice covers ambiguous or multi-intent support requests; it does not measure general out-of-domain detection.

With 20 cases, test clarification on your own traffic. The intervals describe uncertainty in each model's observed rate. This report does not include a paired statistical comparison of the differences between models. This is not a ranking.

See the per-label evidence

Where it fits

Teams that prioritize the lower provider-listed input price and will test whether its task quality meets their routing threshold.

Observed boundary

10 of its 13 observed mistakes were clarification cases routed to a specific team. Estimated at about $0.027 per 1,000 tickets from input tokens only; this is an estimate, not a measured cost.

Provider facts

Provider model ID: @cf/cloudflare/clef-flash
Documented context: 65,536 tokens
Listed input price: $0.09 per 1M input tokens
Source checked: October 9, 2026

Cloudflare documents Clef Flash as a fast hosted 9B multimodal decision model for typed Noul, Choice, and Score questions, with text, structured-data, image, and video state.

Official provider documentation Official pricing source

Concrete observed failures

These are two model-specific mismatches from the frozen held-out results. The case text is fictional.

support-0057Clef Flash

I can sign in and view billing, but downloading invoice INV-905 returns a server error.

Expected: technical
Observed: billing
Status: succeeded

support-0096Clef Flash

For a future 40-seat subscription, can we pay annually by invoice?

Expected: sales
Observed: billing
Status: succeeded

See every observed failure

What this profile cannot answer

It does not estimate performance on your taxonomy, languages, message mix, or escalation policy.
Latency reflects a single disclosed run environment with concurrency one, no retries, and no cache.
Provider aliases may move to newer revisions; record the resolved revision in every evaluation.
Provider input price does not equal total application cost, and this run has no per-attempt billing attribution.