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.

BasismodellAlle Modelle dieser BasisAlle Modelle des Anbieterszum Anbieter

Lesen Sie auch: gpt-oss API-Preise: gpt-oss-120b und gpt-oss-20b bei allen Anbietern

Alle OpenAI-Modelle bei allen Anbietern

Eingabe, $/1M $0.03

Ausgabe, $/1M $0.14

Kontext 131 072

Max. Ausgabe 16 384

Akzeptiert Text

Zuletzt geprüft 04.11.2026 12:00:14 UTC

Quelle —