Models

curl https://api.throughputs.dev/v1/models \
  -H "Authorization: Bearer $THROUGHPUTS_API_KEY"

type Model = {
  id: string;                          // slug, used in `model` field of requests
  name: string;
  provider: {
    id: string;
    name: string;
    slug: string;
  };
  modality: Array<"text" | "image" | "audio" | "embedding" | "code" | "multimodal">;
  capabilities: Array<"streaming" | "function-calling" | "vision" | "json-mode" | "tool-use" | "fine-tuning">;
  contextWindow: number;
  parameters: string | null;           // e.g. "175B"
  pricing: {
    inputPerMillion: number;
    outputPerMillion: number;
    imagePerUnit: number | null;
    unitLabel: string | null;
    free: boolean;
  };
  latencyMs: { p50: number; p95: number } | null;
  throughputTokensPerSecond: number | null;
  qualityScore: number | null;         // 0–10
  popularityRank: number | null;        // 1 = most popular
  status: "active" | "beta" | "deprecated";
};

type ListResponse = {
  object: "list";
  data: Model[];
};

curl https://api.throughputs.dev/v1/models/gpt-4o \
  -H "Authorization: Bearer $THROUGHPUTS_API_KEY"

const response = await fetch("https://api.throughputs.dev/v1/models", {
  headers: { Authorization: `Bearer ${apiKey}` },
});
const { data } = await response.json();

const chatModels = data.filter((m) => m.modality.includes("text"));
const visionModels = data.filter((m) => m.capabilities.includes("vision"));
const cheapModels = data
  .filter((m) => m.pricing.inputPerMillion < 1)
  .sort((a, b) => a.pricing.inputPerMillion - b.pricing.inputPerMillion);