{"data":{"id":"deepinfra/openai/gpt-oss-20b","provider_id":"deepinfra","name":"openai/gpt-oss-20b","modalities":["text"],"context_window":131072,"family":"gpt-oss-20b","status":"active","description":"gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass, optimized for lower-latency inference. The model is trained in OpenAI’s Harmony response format and supports reasoning level configuration, fine-tuning, and agentic capabilities including function calling, tool use, and structured outputs.","architecture":null,"supported_parameters":["chat","reasoning_effort","reasoning"],"created_at":null,"canonical_slug":"openai/gpt-oss-20b","catalog_source":"deepinfra","host_id":"deepinfra","host_name":"DeepInfra","creator_id":"openai","creator_name":"Openai","cheapest_provider":null,"cheapest_provider_name":null,"cheapest_price":null,"savings_pct":null,"savings_basis":null,"provider":{"id":"deepinfra","name":"DeepInfra","website":"https://deepinfra.com","categories":["text","image","audio","video"],"docs_url":"https://deepinfra.com/pricing","notes":"Secondary live feed from DeepInfra's public models API. Prices are DeepInfra's rates, not the origin lab's. Does not block the OpenRouter sync.","role":"host"},"prices":[{"id":60233,"model_id":"deepinfra/openai/gpt-oss-20b","unit":"token","input_per_1m_usd":0.03,"output_per_1m_usd":0.14,"flat_usd":null,"currency":"USD","effective_at":"2026-10-08","source_url":"https://deepinfra.com/openai/gpt-oss-20b","notes":"DeepInfra public models API. Token and character rates are already USD per 1M. Per-image and per-second rates are stored as flat USD. Discount, when present, is not applied.","last_verified_at":"2026-10-08T18:23:40.278Z","source":"deepinfra","confidence":"official","cache_read_per_1m_usd":null,"cache_write_per_1m_usd":null,"cache_write_1h_per_1m_usd":null,"pricing_detail":null,"long_context_override":false}]}}