Kimi K3 (2.8T parameters, 1M context)

Open weights & AI Act facts

Factual characteristics of this model's weights — stated, not certified. Under the EU AI Act, whether a deployment is compliant depends on the model, your role and your use case.

Open weights

Weights public — use governed by the licence above

AI Act class

GPAI — general-purpose AI

Systemic risk

Not assessed by us — training compute not published, or too close to the 10²⁵ FLOP threshold to call

Data retention

None at Frontière AI; provider-side retention follows Modal's own policy (not verified by us)

Frontière Verified

Our 6-criteria checklist for this model. Each criterion is assessed independently — a checkmark means we ran, recorded and passed the check; an empty circle means not confirmed yet.

Compliance evidence

Provider facts for this model's serving path, as documented by Frontière AI; the retention row quotes each provider's own documentation where it exists. Dates reflect the most recent documented audit of each fact.

Last verified: 2026-09-11

Provider / operator

Modal (US company)

Country

United States

Ownership / control

US-controlled — CLOUD Act applies even in EU regions

Jurisdiction

US (default serving region observed outside the EU)

Deployment

Managed serverless (third party)

Data retention

None at Frontière AI; provider-side retention follows Modal's own policy (not verified by us)

Fallback route

OpenRouter (US company, New York) — None at Frontière AI; provider-side retention follows the policy of whichever provider OpenRouter picks (not verified by us)

Frontière AI retains neither prompts nor completions from API calls; processing is structured to fit a short DPA (Art. 28 GDPR). DPA and subprocessor documentation available on request.

What its label means

This model is served through infrastructure controlled by a US company, which remains subject to the US CLOUD Act even when it runs in an EU region. We label it “fast access”, never sovereign, and it is flagged in the API response before you call it — so the choice is yours, made with full information. If Modal fails (429, 5xx or unreachable), the call automatically falls back to OpenRouter, which then picks the final provider from its own pool — a pool we do not control. The X-Routed-Provider header on every response names the route used.

Sovereignty Score — 0/100

Deterministic 0–100 score computed from the five factors below — ownership, hosting jurisdiction, cloud-provider control, data retention, deployment control. Full methodology: How Frontière AI determines sovereignty →

Operating-company ownership

0/40

Hosting jurisdiction (observed)

0/30

Cloud-provider control

0/15

Data retention

0/10

Deployment control

0/5

Effective provider assessed: modal.

Calling it

The API is OpenAI-compatible. Change base_url and the key, then use this model id:

client = OpenAI(
    base_url="https://getfrontiereai.eu/api/v1",
    api_key=os.environ["FRONTIERE_KEY"],
)

reply = client.chat.completions.create(
    model="kimi-k3",
    messages=[{"role": "user", "content": "Bonjour"}],
)

One key covers the whole catalog — you switch models by changing a single string.

Pricing

The price you see is the price debited from your prepaid balance, margin and VAT included — not an infrastructure cost. No subscription: top up from €10; calls are refused once the balance reaches zero, and a call already running can take it slightly below — deducted from your next top-up.

What will it cost you?

Pick a workload or type your own token counts: the estimate uses the prices above, margin and VAT included.

FAQ

Articles about this model

Other models in the catalog