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.
Qwen: Qwen3 Next 80B A3B Thinking
qwen/qwen3-next-80b-a3b-thinkingTry Qwen: Qwen3 Next 80B A3B Thinking 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
Qwen: Qwen3 Next 80B A3B Thinking at a glance
- Input price is 67% lower than the median of 434 Text models.
- Context window of 262,144 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 $0.00262.
About Qwen: Qwen3 Next 80B A3B Thinking
Qwen: Qwen3 Next 80B A3B Thinking is a text model from Qwen, part of the Qwen3-Next line. It is a reasoning-first chat model, which means it produces structured thinking traces by default rather than answering immediately. The provider released it on 11 September 2025, and its knowledge cutoff is 30 September 2025.
The model accepts text input and returns text output. Its context window is 262,144 tokens, and the maximum output is 235,929 tokens, so long inputs and long generations are both possible within those bounds. Because it is reasoning-first, responses typically include a thinking trace alongside the final answer, and the include_reasoning and reasoning parameters let you control how that trace is handled in the response. It is built for hard multi-step problems rather than quick conversational replies.
That design points to a fairly specific set of use cases. The provider describes it as suited to math proofs, code synthesis and debugging, logic problems, and agentic work. In practice, you would reach for it when a task needs several dependent steps, such as working through a derivation, tracing a bug across a codebase, or planning and executing a sequence of tool calls. It is less obviously a fit for simple, latency-sensitive chat, where a non-reasoning model would usually be the more direct choice.
Through api14 you call it with the model id "qwen/qwen3-next-80b-a3b-thinking" over the OpenAI-compatible API, so existing OpenAI client libraries work with a change of base URL and model id. The supported request parameters are frequency_penalty, include_reasoning, logprobs, max_tokens, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs and top_p. Tool calling and structured outputs are both available, which matters for the agentic and code-oriented work described above, and logprobs plus top_logprobs give you token-level visibility when you need it.
A few practical notes follow from the facts. Input and output are text only, so image or audio tasks need a different model. The large context window and output limit suit long documents and extended reasoning, but the thinking traces add output tokens you may want to account for when setting max_tokens. If your task is short and does not benefit from step-by-step reasoning, a model without a reasoning-first design may be the simpler fit.
How to call the Qwen: Qwen3 Next 80B A3B Thinking 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 "qwen/qwen3-next-80b-a3b-thinking".
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": "qwen/qwen3-next-80b-a3b-thinking",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
Supported parameters
frequency_penaltyinclude_reasoninglogprobsmax_tokenspresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_pSpecifications
- Context length
- 262,144
- Max output
- 235,929
- Input
- text
- Output
- text
- Knowledge cutoff
- 2025-09-30
Requests on api14 · last 7 days
Qwen: Qwen3 Next 80B A3B Thinking: frequently asked questions
- How much does the Qwen: Qwen3 Next 80B A3B Thinking API cost?
- Qwen: Qwen3 Next 80B A3B Thinking costs Input $0.52 / 1M tokens. It is charged from your USD balance only for successful requests; failed requests are free.
- What is the context window of Qwen: Qwen3 Next 80B A3B Thinking?
- Qwen: Qwen3 Next 80B A3B Thinking reads up to 262,144 tokens per request: prompt, history and documents together. It writes up to 235,929 output tokens per response.
- What can Qwen: Qwen3 Next 80B A3B Thinking do?
- Qwen: Qwen3 Next 80B A3B Thinking accepts text and returns text. It supports tool calling, structured JSON output, reasoning effort control.
- How do I call Qwen: Qwen3 Next 80B A3B Thinking from my code?
- Point any OpenAI SDK at https://api14.com/api/v1 and set the model to "qwen/qwen3-next-80b-a3b-thinking". No other code changes. Copy a ready example from the code panel on this page.
- What are alternatives to Qwen: Qwen3 Next 80B A3B Thinking?
- Comparable Text models at a similar price: StepFun: Step 3.7 Flash, TheDrummer: Skyfall 36B V2, Qwen: Qwen3.6 Flash. Each runs through the same API key, so switching is one string.
Qwen: Qwen3 Next 80B A3B Thinking community
New topicThis community has no topics yet. Open the first one about Qwen: Qwen3 Next 80B A3B Thinking.