FreeModel
A third-party model catalog (allaimodel.com) lists FreeModel (freemodel.dev) as a reseller of GPT-5.4 mini at $0.75 per 1M input tokens and $4.50 per 1M output tokens, with cache reads priced at $0.075 and cache writes at $0.75 per 1M tokens. FreeModel is shown alongside roughly 14 other resellers (Poe, Impossibl, Cros The same catalog entry reports GPT-5.4 mini on FreeModel ships with a 400K-token context window and 128K-token maximum output, accepts text and image inputs while emitting text outputs, supports reasoning and tool calling, and has an August 2025 knowledge cutoff with a release date of 2026-03-17. This gives developers