We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
OpenAI: gpt-oss-20b (batch)
openai/gpt-oss-20b:batchTry OpenAI: gpt-oss-20b (batch) now
Runs on your balance at the prices above. New accounts start with free usage.
One API. Every model.
Session cost $0.00
Settings
Advanced parameters
OpenAI: gpt-oss-20b (batch) at a glance
- Input price is 95% lower than the median of 434 Text models.
- Context window of 131,072 tokens, enough for long documents and multi-turn conversations.
- Supports tool (function) calling for agents and integrations.
- Supports structured JSON output with a schema.
- Supports reasoning effort control.
- A typical request (1,000 input / 500 output tokens) costs about $0.000280.
About OpenAI: gpt-oss-20b (batch)
OpenAI: gpt-oss-20b (batch) is an open-weight text model from OpenAI, released under the Apache 2.0 license. It is a 21B parameter model built on a Mixture-of-Experts (MoE) architecture with 3.6B active parameters per forward pass. The batch variant is served through api14 as a text generation endpoint, and the model id you use in requests is openai/gpt-oss-20b:batch.
The model accepts text input and produces text output. It has a context window of 131072 tokens and a maximum output of 117964 tokens, with a knowledge cutoff of 2024-06-30. Because it is a reasoning-capable model, requests can enable reasoning output and control how much reasoning effort is applied. It also supports structured outputs and response formats, which makes it suitable for tasks where the shape of the returned text matters, not just its content.
Those capabilities point to a fairly wide set of practical uses. Long context handling suits document summarization, multi-file code review, and question answering over large reference material. Reasoning support helps with multi-step problems, planning, and analysis where the model benefits from working through intermediate steps. Tool calling and structured outputs fit agent-style workflows, data extraction into fixed schemas, and pipelines that need machine-readable responses. The batch variant is aimed at workloads where throughput matters more than immediate interactivity, such as offline generation, bulk classification, and dataset enrichment.
To call it through api14, use the OpenAI-compatible chat completions interface with the model id openai/gpt-oss-20b:batch. The request body accepts frequency_penalty, include_reasoning, logit_bias, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, and top_p. If you already use an OpenAI client library, you can point it at the api14 base URL and keep your existing request structure, adjusting only the model id and any parameters you want to set.
Keep the knowledge cutoff in mind: the model does not know events after 2024-06-30, so anything time-sensitive may need retrieval or a different model. Input is text only, so image or audio tasks fall outside its scope. If your workload needs low-latency interactive responses rather than batch processing, a non-batch deployment may be a better fit.
How to call the OpenAI: gpt-oss-20b (batch) API
- Create an API key in your dashboard.
- Install the official openai SDK for your language.
- Set baseURL to https://api14.com/api/v1 and model to "openai/gpt-oss-20b:batch".
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "openai/gpt-oss-20b:batch",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
Supported parameters
frequency_penaltyinclude_reasoninglogit_biasmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_pSpecifications
- Context length
- 131,072
- Max output
- 117,964
- Input
- text
- Output
- text
- Knowledge cutoff
- 2024-06-30
Requests on api14 · last 7 days
OpenAI: gpt-oss-20b (batch): frequently asked questions
- How much does the OpenAI: gpt-oss-20b (batch) API cost?
- OpenAI: gpt-oss-20b (batch) costs Input $0.08 / 1M tokens. It is charged from your USD balance only for successful requests; failed requests are free.
- What is the context window of OpenAI: gpt-oss-20b (batch)?
- OpenAI: gpt-oss-20b (batch) reads up to 131,072 tokens per request: prompt, history and documents together. It writes up to 117,964 output tokens per response.
- What can OpenAI: gpt-oss-20b (batch) do?
- OpenAI: gpt-oss-20b (batch) accepts text and returns text. It supports tool calling, structured JSON output, reasoning effort control.
- How do I call OpenAI: gpt-oss-20b (batch) from my code?
- Point any OpenAI SDK at https://api14.com/api/v1 and set the model to "openai/gpt-oss-20b:batch". No other code changes. Copy a ready example from the code panel on this page.
- What are alternatives to OpenAI: gpt-oss-20b (batch)?
- Comparable Text models at a similar price: Mistral: Mistral Small 3, Meta: Llama 3.1 8B Instruct, IBM: Granite 4.0 Micro. Each runs through the same API key, so switching is one string.
OpenAI: gpt-oss-20b (batch) community
New topicThis community has no topics yet. Open the first one about OpenAI: gpt-oss-20b (batch).