generated: '2026-08-26'
method: searched
source: https://vynaris.com/pricing
docs:
- https://vynaris.com/pricing
- https://vynaris.com/enterprise
- https://vynaris.com/uncensored-models
- https://api.vynaris.com/agents.md
currency: USD
plan_count: 5
billing_model: >-
  Prepaid credit. Two ways to fund it: one-time top-ups, or a fixed monthly
  plan whose charge converts to API credit. Usage is deducted per request at
  the upstream provider's list price plus a convenience fee, or at published
  retail rates for Vynaris-hosted models. No usage overages - requests stop at
  a zero balance.
free_tier: false
free_tier_note: >-
  Signup includes no free credit. Starting balance is $0 and a human must add
  at least $50 of prepaid credit before the first billable request. A
  programmatically created account authenticates but returns 402 until funded.
plans:
- name: Pay-as-you-go (router)
  price: 0
  billing_period: none
  model: usage-based on prepaid credit
  unit_price: upstream provider list price x 1.03, dropping to x 1.01 once monthly usage passes $500
  fee_note: >-
    A $500 top-up activates the 1% rate immediately for that calendar month.
  funding: one-time top-ups of $50, $100, $250 or $500
  minimum_first_funding: 50
  included_quota: none
  overage: none - requests stop when available balance reaches zero
  limits:
  - no overage charges
  - optional per-key spend caps
  - credit does not expire silently
  source: https://vynaris.com/pricing
- name: Vynaris S
  price: 5
  currency: USD
  billing_period: monthly
  included_quota: $5 in API credit every month
  overage: none
  rollover: unused credit rolls over
  features:
  - full model pool + automatic routing
  - OpenAI-compatible API
  - privacy-restricted providers by default
  - no training by Vynaris
  - per-request receipts + CSV ledger
  - top up anytime; no overage charges
  - reliability refunds
  source: https://vynaris.com/pricing
- name: Vynaris L
  price: 15
  currency: USD
  billing_period: monthly
  included_quota: $15 in API credit every month
  overage: none
  rollover: unused credit rolls over
  features:
  - full model pool + automatic routing
  - OpenAI-compatible API
  - privacy-restricted providers by default
  - no training by Vynaris
  - per-request receipts + CSV ledger
  - top up anytime; no overage charges
  - reliability refunds
  source: https://vynaris.com/pricing
- name: Vynaris XL
  price: 50
  currency: USD
  billing_period: monthly
  included_quota: $50 in API credit every month
  overage: none
  rollover: unused credit rolls over
  features:
  - private uncensored model access (qualifies via the $50 credit purchase)
  - full model pool + automatic routing
  - OpenAI-compatible API
  - privacy-restricted providers by default
  - no training by Vynaris
  - per-request receipts + CSV ledger
  - top up anytime; no overage charges
  - reliability refunds
  source: https://vynaris.com/pricing
- name: Enterprise (self-hosted)
  price: 5000
  currency: USD
  billing_period: annual
  contact_sales: true
  target: teams spending $5k-100k/month on LLM APIs
  included: installation, updates, and security patches
  model: >-
    The gateway is installed on the customer's own infrastructure and operated
    by them, routing through the customer's own provider accounts at their own
    rates. Vynaris is never in the compute bill.
  source: https://vynaris.com/enterprise
hosted_model_rates:
  note: >-
    Vynaris-hosted reduced-refusal models are billed at published retail token
    rates rather than upstream-cost-plus-fee, and are never substituted by the
    router. A successful purchase of at least $50 in API credit qualifies an
    account; access remains available as that balance is used.
  models:
  - id: vynaris/qwen3.6-35b-a3b-uncensored
    input_per_million_usd: 1.00
    output_per_million_usd: 5.00
  - id: vynaris/qwen3.8-27b-uncensored
    input_per_million_usd: 1.00
    output_per_million_usd: 7.00
  - id: vynaris/deepseek-v4-flash-uncensored
    input_per_million_usd: 2.00
    output_per_million_usd: 11.00
  context_window: up to 128K API context
  source: https://vynaris.com/pricing
announced_not_yet_priced_as_a_plan:
- name: Kimi K3 founding reservation
  input_per_million_usd: 1.50
  output_per_million_usd: 7.00
  status: reservation from $50, pre-launch
  source: https://vynaris.com/kimi-k3-inference
payments:
  merchant_of_record: third-party payment partner (named as Dodo Payments in agents.md; the pricing page says only "our merchant-of-record payment partner")
  methods_count: 26
  methods_note: >-
    Cards, Apple/Google/Amazon Pay, Cash App, Klarna, Afterpay/Clearpay, UPI,
    iDEAL, Bancontact, EPS, Multibanco, Revolut Pay, Satispay, Billie, Pix,
    WeChat Pay, Kakao Pay, Naver Pay, Przelewy24, BLIK, Zip, and stablecoins
    (USDC, USDP, USDG).
  refunds: unused credits refundable on request minus processor fees; promotional credit is not refundable
  source: https://vynaris.com/pricing
machine_readable_pricing:
  published: false
  note: >-
    GET /v1/models is documented to return the model catalog with retail
    prices, context and paid-credit qualification - but it requires a funded
    key (401 anonymously on 2026-08-26), so no anonymous machine-readable
    price feed exists. https://vynaris.com/models is a human pricing table.