readable AI benchmarks (simplified)

GET/api/models

Catalog endpoint for every explorer dataset. Omit dataset or use ?dataset=language for the existing language response, including canonical Quality Score and Quality Coverage. Use ?dataset=image or ?dataset=video for the Image and Video catalogs (Quality remains qualityElo), and ?dataset=all for all three named envelopes. Image and video rows may also include externalBenchmarks for exact LM Arena Score matches. That field is additive evidence and does not change qualityEloor default ranking. Envelope source / stale describe the primary Artificial Analysis media feed only. LM Arena live / cached / snapshot status lives in additive externalBenchmarkSources; a live AA catalog does not imply a live Arena fetch. It is a separate comparison group, not a ranking; use /api/recommend for ranked language selection.

curl -sS "https://models.deggo.fyi/api/models?dataset=image"
curl -sS "https://models.deggo.fyi/api/models?dataset=video"
curl -sS "https://models.deggo.fyi/api/models?dataset=all"

Language responses keep the existing source, updatedAt, models, and message envelope. Image and video responses add sourceUrl and stale for the primary catalog, plus externalBenchmarkSources for attached external boards. The all-dataset response returns those complete envelopes under datasets.language, datasets.image, and datasets.video. Exact language DeepSWE attachments expose deepSwe.meanAgentSteps, deepSwe.meanCostUsd, deepSwe.meanOutputTokens, and deepSwe.meanDurationSeconds (each number or null); the Graph reads that same raw steps value. Every language models[] row includes:

• slug: canonical model slug
• name: display name or null
• provider: creator name
• qualityValue: current Quality Score, number 0-100 or null when ineligible
• qualityCoverage: 1 when required AA inputs are valid, otherwise null
• attemptRate: Correct + Incorrect, number 0-1 or null
• priceValue: current Value Score, number 0-100 or null

Existing catalog pricing, benchmark, modality, and capability fields remain unchanged. Missing Quality is returned as null, never 0, and no catalog rows are omitted.

Example Request

curl -sS "https://models.deggo.fyi/api/models"

GET/api/recommend

ranked pick for a constraint set. Default goal=quality, limit=10 (maximum 30). Use this for selecting models, not mirroring the full catalog; bulk catalog consumers should use /api/models. Default body also includes price[] and cheap[]. read is the first field after source.

Query

Unknown keys return 400.

Response fields

models[] row

Example Request

curl -sS "https://models.deggo.fyi/api/recommend"
curl -sS "https://models.deggo.fyi/api/recommend?q=grok"
curl -sS "https://models.deggo.fyi/api/recommend?goal=cheap"

from urllib.request import Request, urlopen

req = Request(
    "https://models.deggo.fyi/api/recommend",
    headers={"Accept": "application/json"},
)
with urlopen(req) as response:
    print(response.read().decode())

Example Response

{
  "source": "recommend",
  "read": "models[] is the requested ranking. default goal=quality. price[] is quality per dollar, not cheapest USD. cheap[] is lowest blendedPrice above 0. lookup one model: GET /api/recommend?q=NAME. full catalog with Quality values: GET /api/models. HTML is a graph view, not the ranking.",
  "goal": "quality",
  "updatedAt": "<from models feed, else null>",
  "catalogSource": "<live|cached|snapshot|null>",
  "catalogUpdatedAt": "<original models timestamp or null>",
  "catalogStale": true,
  "catalogMessage": "<models message or null>",
  "note": "catalog provenance. HTML is a graph view, not the ranking. do not scrape /.",
  "filters": { "goal": "quality", "limit": 10 },
  "count": 10,
  "models": [
    {
      "slug": "",
      "name": "",
      "provider": "",
      "reasoning": true,
      "openWeights": false,
      "contextWindowTokens": 0,
      "inputModalities": [],
      "outputModalities": [],
      "blendedPrice": 0,
      "intelligence": 0,
      "coding": 0,
      "agentic": 0,
      "correct": 0,
      "incorrect": 0,
      "qualityValue": 0,
      "qualityCoverage": 1,
      "attemptRate": 0,
      "priceValue": 0,
      "why": "short sentence from actual numbers"
    }
  ],
  "price": [],
  "cheap": []
}

Errors

GET/api

Machine index. Fetch this first if query keys are unknown. Field use explains models[] vs price[] vs cheap[].

Example Request

curl -sS "https://models.deggo.fyi/api"