Model Library/OpenAI: GPT OSS 20B

OpenAI: GPT OSS 20B

openai/gpt-oss-20b

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 and deployability on consumer or single-GPU hardware. 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.

Características

API serverless

Documentación

openai/gpt-oss-20b is available via Novita's serverless API, where you pay per token. There are several ways to call the API, including OpenAI-compatible endpoints with exceptional reasoning performance.

Implementaciones bajo demanda

Documentación

On-demand deployments allow you to use openai/gpt-oss-20b on dedicated GPUs with high-performance serving stack with high reliability and no rate limits.

Serverless disponible

Ejecuta consultas de inmediato, paga solo por el uso

Entrada$0.04 / M Tokens

Salida$0.15 / M Tokens

Usa los siguientes ejemplos de código para integrarte con nuestra API:

1from openai import OpenAI
2
3client = OpenAI(
4    api_key="<Your API Key>",
5    base_url="https://api.novita.ai/openai"
6)
7
8response = client.chat.completions.create(
9    model="openai/gpt-oss-20b",
10    messages=[
11        {"role": "system", "content": "You are a helpful assistant."},
12        {"role": "user", "content": "Hello, how are you?"}
13    ],
14    max_tokens=32768,
15    temperature=0.7
16)
17
18print(response.choices[0].message.content)

Información

Proveedor

OpenAI

Cuantización

fp4

Funcionalidad compatible

Longitud del contexto

131072

Salida máxima

32768

Serverless

Compatible

Structured Output

Compatible

Reasoning

Compatible

Capacidades de entrada

text, image

Capacidades de salida

text

Todo lo que necesitas para crear IA de producción.

Más de 200 modelos, GPUs bajo demanda y entornos de ejecución seguros para agentes, unificados bajo una API. Gratis para empezar, escala a medida que creces.