Qwen: Qwen3 Next 80B A3B Instruct

Text input Text output
Author's Description

Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual use, while remaining robust on alignment and formatting. Compared with prior Qwen3 instruct variants, it focuses on higher throughput and stability on ultra-long inputs and multi-turn dialogues, making it well-suited for RAG, tool use, and agentic workflows that require consistent final answers rather than visible chain-of-thought. The model employs scaling-efficient training and decoding to improve parameter efficiency and inference speed, and has been validated on a broad set of public benchmarks where it reaches or approaches larger Qwen3 systems in several categories while outperforming earlier mid-sized baselines. It is best used as a general assistant, code helper, and long-context task solver in production settings where deterministic, instruction-following outputs are preferred.

Key Specifications
Cost
$$$$
Context
262K
Parameters
80B
Released
Sep 11, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Presence Penalty Frequency Penalty Tools Tool Choice Top Logprobs Top P Seed Logprobs Max Tokens Temperature Stop Min P Logit Bias
Features

This model supports the following features:

Tools
Performance Summary

Qwen3-Next-80B-A3B-Instruct, created on September 11, 2025, demonstrates strong performance as an instruction-tuned chat model. It consistently ranks in the top tier for speed, placing in the 76th percentile across six benchmarks, indicating its efficiency in generating responses. While its pricing is moderate, falling into the 40th percentile, the model exhibits exceptional reliability with a perfect 100% success rate, ensuring consistent and usable outputs without technical failures. In terms of benchmark performance, the model achieved perfect accuracy in both General Knowledge and Ethics, often being the most accurate at its price point and speed. It also performed very well in Email Classification (99.0% accuracy) and Coding (92.0% accuracy). Its Reasoning capabilities are strong at 88.0% accuracy, and it stands out as the most accurate among models of comparable speed in this category. The primary area for improvement appears to be Instruction Following, where it achieved 63.0% accuracy. Overall, Qwen3-Next-80B-A3B-Instruct is a robust model well-suited for production environments requiring fast, stable, and reliable instruction-following, particularly for complex tasks, RAG, and agentic workflows.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.3
Completion $0.3

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Hyperbolic
Hyperbolic | qwen/qwen3-next-80b-a3b-instruct-2509 262K $0.3 / 1M tokens $0.3 / 1M tokens
Alibaba
Alibaba | qwen/qwen3-next-80b-a3b-instruct-2509 131K $0.5 / 1M tokens $2 / 1M tokens
Novita
Novita | qwen/qwen3-next-80b-a3b-instruct-2509 65K $0.15 / 1M tokens $1.5 / 1M tokens
Chutes
Chutes | qwen/qwen3-next-80b-a3b-instruct-2509 262K $0.147 / 1M tokens $0.587 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by qwen