GPT OSS 20B · deepinfra

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.

Modelo baseTodos los modelos de esta baseTodos los modelos del proveedoral proveedor

Lea también: Precios de la API de gpt-oss: gpt-oss-120b y gpt-oss-20b en todos los proveedores

Todos los modelos de OpenAI en todos los proveedores

Entrada, $/1M $0.03

Salida, $/1M $0.14

Contexto 131 072

Salida máx. 16 384

Acepta texto

Última comprobación 04.11.2026 12:00:14 UTC

Fuente —