Apipriceindex

Search the API Price Index for LLM API endpoints by name, id or provider. Returns matching models with their `id` (use it with get_price / estimate_cost / offers_table), provider, context window and USD prices per million tokens. Optional filters: provider slug (e.g. 'openai'), minimum context window, open-weights only, max_price (input + output USD per million tokens), capabilities (any of tools, reasoning, vision, json — as declared by the host), min_speed (output tokens per second, OpenRouter median over 30 min). An empty query lists everything that passes the filters.

Get the verified price of one model (USD per million tokens). Returns input/output/cached-input prices with verification date, cross-check confidence and source URL. Prices are re-checked daily, from each provider's own price list or from OpenRouter's per-provider data; source_url says which. For open-weight models, served.precision, served.context and served.max_output say what this host serves, each with its state (published, not_stated, not_read, to_check), source and date. Data: apipriceindex.com, CC BY 4.0.

Estimate the monthly USD bill of a workload and rank models by cost. Give monthly volumes in millions of tokens (input and output) and an optional share of input served from cache (0-100). Ranks the given model ids — or the whole catalog — from cheapest to most expensive. Cost only: this tool does not rank model quality.

List today's LLM API deals: models free to call right now, discounts, free credits and free tiers. kind: all, free, discount or credits. ending_within_days: only deals whose host-published end date falls within that many days (e.g. 2 for "what ends tomorrow?"). Every deal carries its end date ("no end date" when the host gives none), its official source and the date it was checked. Free models are read every morning from OpenRouter, host pages twice a day.

Every host serving one model, as a table with the columns you choose. model: a model id, base model or name (find it with search_models). columns: any of input, output, cache, discount, speed, first_token, uptime, precision, context, max_output, verified_at (default: input, output, cache, speed, uptime, verified_at). Rows are sorted cheapest first. Speed and uptime come from OpenRouter with their window; prices from each host's source; precision, context and max_output from the host's own catalogue or docs where its terms allow, otherwise from OpenRouter, and an empty one comes with <column>_state saying why.

Compare two to four models side by side. For each: lowest input, output and cached-input price with the provider that gives it, number of providers, fastest provider (OpenRouter median), context window, license, Epoch Capabilities Index score; `best` names the winner of each row. Give monthly volumes (millions of tokens) to add the monthly cost at each model's cheapest provider.

Cheapest models that reach a given intelligence level. Level = Epoch Capabilities Index score (Epoch AI, CC BY 4.0): pass min_score, or as_good_as a model to use its score. Returns the cheapest models at or above it (lowest input + output price among their providers), optionally open weights only.

Price drops and rises over the last N days, and the models released in that time. direction: all, drops or rises. Each change compares the last recorded price before the period with today's (input + output per million tokens). New models come with their cheapest provider.