# Pricing transcribed from the provider's own page — see provenance. Numbers are quoted, not derived.
generated: '2026-10-07'
method: searched
source: https://www.modular.com/pricing
sources:
- https://www.modular.com/pricing
provenance:
  tool: all/0-working/economics/harvest_pricing.py
  run: 20261007T104522
  grounding: every price_text was found verbatim in the fetched page text
audience: developer-api
pricing_basis: per-call
delivery:
  delivery: open-core
  method: page
  read: '2026-10-07'
  evidence: The Self-Hosted Community Edition is completely free and open source.
  free_edition: Community Edition
  managed_tier: Modular Cloud
  open_source_license: Modular Community License
published: true
note: Page lists per-token hosted model API prices ($/1M input, output, cache hit) for ~20 models and
  $/1K @1MP image model prices, plus four editions (Modular Cloud per token or GPU hour, BYOC per minute,
  Enterprise custom, free self-hosted Community Edition). Edition tier prices themselves are not numeric;
  volume/custom pricing via 'Talk to Sales'.
plan_count: 7
currency: USD
plans:
- name: Modular Cloud
  price_text: PER TOKEN OR GPU HOUR
  currency: USD
  period: per-use
  unit: token or GPU hour
  included_quota: Shared & Dedicated Endpoints; start by testing shared endpoints for free
- name: Bring Your Own Cloud
  price_text: PAY PER MINUTE
  period: per-use
  unit: minute of reserved GPU capacity
  included_quota: Everything in Dedicated Endpoint, plus deployment in your cloud or on-premise
- name: Enterprise
  price_text: CUSTOM ENGAGEMENT
  period: none
  included_quota: SOTA inference on any GPU vendor, custom kernels, community support
- name: Self-Hosted
  price_text: Free
  period: none
  included_quota: Community Edition, free forever with no usage fees
- name: DeepSeek V4 (hosted endpoint)
  price_text: '1.74'
  price: 1.74
  currency: USD
  period: per-use
  unit: $/1M input tokens
  included_quota: Output 3.48 per 1M; Cache Hit 0.145 per 1M
- name: GPT OSS 120B (hosted endpoint)
  price_text: '0.1'
  price: 0.1
  currency: USD
  period: per-use
  unit: $/1M input tokens
  included_quota: Output 0.5 per 1M; Cache Hit N/A
- name: FLUX.2-klein-4B (image)
  price_text: '1'
  price: 1
  currency: USD
  period: per-use
  unit: $/1K @ 1MP input
  included_quota: Output 1 per 1K @ 1MP
x-canonical:
  generated: '2026-10-08'
  vocabulary: all/0-working/economics/plans-canonical.yml
  models:
  - hybrid
  - metered
  - subscription
  units:
  - compute
  - token
  plans:
  - name: Modular Cloud
    access: other
    model: metered
    model_how: derived
    units:
    - token
    periods:
    - other
  - name: Bring Your Own Cloud
    access: other
    model: metered
    model_how: derived
    units:
    - compute
    periods:
    - other
  - name: Enterprise
    access: other
    model: unstated
    model_how: derived
    units: []
    periods:
    - none
  - name: Self-Hosted
    access: other
    model: unstated
    model_how: derived
    units: []
    periods:
    - none
  - name: DeepSeek V4 (hosted endpoint)
    access: other
    model: hybrid
    model_how: derived
    units:
    - token
    periods:
    - other
  - name: GPT OSS 120B (hosted endpoint)
    access: other
    model: hybrid
    model_how: derived
    units:
    - token
    periods:
    - other
  - name: FLUX.2-klein-4B (image)
    access: other
    model: subscription
    model_how: derived
    units:
    - other
    periods:
    - other