Unified AI API Model Catalog — 300+ Models Behind One OpenAI-Compatible API

image2.5 from $0.0085 per 1K image · Seedance 2.5 from $0.0961/sec · cached LLM input from $0.40/M — one OpenAI-compatible endpoint at https://api.apimart.ai/v1, no monthly plan required. (observed 2026-09-17)

Get an API key · Live pricing · Model page

Why teams route through APIMart

• One key, entire catalog. The same https://api.apimart.ai/v1 base URL and Authorization header reach the whole catalog behind one key and 300+ other image, video and language models — switch the model field, not your client.
• $1 minimum, pay as you go. No subscription and no prepaid plan to size up front: top up from $1 and spend it on calls. There is no free quota to burn through first, so the price in this table is the price you pay.
• The charge comes back in the response. Every call reports the amount billed (cost / credits_cost), so a spend number is read per call instead of guessed at month end.
• Async by design. Submit, take the task_id, poll GET /v1/tasks/{id} — batching and retries are ordinary queue work, not a bespoke integration.

A unified AI API needs one place that answers "which model ids exist, what modality are they, and what unit do they bill in?". This repository is that place: a machine-readable catalog of the models reachable through a single OpenAI-compatible base URL, regenerated every day from the public pricing payload.

data/models.json — one record per model:

{
  "id": "gpt-image-2.5-ext",
  "name": "gpt-image-2.5-ext",
  "alias": "gpt-image-2.5-ext",
  "specification": "image",
  "unit": "usd_per_image",
  "prices": {
    "flare@1K":    { "list": 0.010625, "effective": 0.0085 },
    "flare@2K":    { "list": 0.0175,   "effective": 0.014 },
    "flare@4K":    { "list": 0.02625,  "effective": 0.021 }
  }
}

Headline routes per modality (the full list, with every model id, is in CATALOG.md):

python examples/lookup.py --spec image --sort price | head            # cheapest image routes first
python examples/lookup.py --grep nano-banana                          # every Nano Banana record
python examples/lookup.py --id gpt-image-2.5-ext                      # one record, pretty-printed
python examples/lookup.py --models                                     # just the model ids, for a dropdown or a test fixture

The lookup script reads the local JSON only — no network, no API key — so it is safe inside CI and inside an agent loop.

• Diffable. data/models.json is committed, so a new model or a price move shows up as a git diff with a timestamp.
• Guess-free. Display names and model ids diverge (Nano Banana Pro vs gemini-3-pro-image-preview), and a catalog is where that mapping lives instead of in someone's notebook.
• Budget-ready. Every record carries its billing unit, so a cost model can pick the right formula before the first paid request.

Linear at the observed per-unit rate, no volume discount assumed. Snapshot 2026-09-17; re-check the live table before committing a budget.

What is a unified AI API? One base URL, one credential and one billing surface in front of many models. The unified part is the envelope and the credential, not the models: each provider keeps its own parameters, limits and quirks.

How do I know which models are available right now? Read data/models.json in this repository (or CATALOG.md) rather than trusting a model list from a blog post. The CI refresh commits a diff whenever the catalog moves.

Do model ids and display names match? Often not. Image models are the worst offenders: the display name may read Nano Banana Pro while the API expects gemini-3-pro-image-preview. Both are in the record — use id in requests and name in documentation.

Can I use this catalog to estimate cost? Yes, at the unit level: each record carries the billing unit and the price cells. Multiply by your volume (images, seconds, tokens) and compare against a completed task's cost field for reconciliation.

• unified ai api
• ai api model catalog
• one api key many models
• ai api aggregator
• llm api list
• image generation api pricing
• openai compatible api

Start with $1. Get an API key → check live pricing → open the whole catalog behind one key in the model library. The first call is three steps: submit, poll task_id, read the charged amount off the response.

Outbound APIMart links are minted through the promo link API; hand-made tracking parameters are rejected by tools/check_links.py in CI.

APIMart is the service whose public pricing page supplies this snapshot; the repository is published to document that catalog, not to claim official status. Model names, prices and documentation belong to their respective owners, and relayed routes are third-party relay endpoints rather than first-party vendor endpoints.

README.md          catalog overview, coverage and schema
CATALOG.md         every model id as a readable table (regenerated in CI)
data/models.json   machine-readable catalog
tools/catalog.py   extractor + table builder
tools/check_links.py  attribution guard
examples/lookup.py    offline query helper
.github/workflows/    daily refresh + validation

MIT — see LICENSE.