Qwen: Qwen3 30B A3B

Text input Text output
Author's Description

Qwen3, the latest generation in the Qwen large language model series, features both dense and mixture-of-experts (MoE) architectures to excel in reasoning, multilingual support, and advanced agent tasks. Its unique ability to switch seamlessly between a thinking mode for complex reasoning and a non-thinking mode for efficient dialogue ensures versatile, high-quality performance. Significantly outperforming prior models like QwQ and Qwen2.5, Qwen3 delivers superior mathematics, coding, commonsense reasoning, creative writing, and interactive dialogue capabilities. The Qwen3-30B-A3B variant includes 30.5 billion parameters (3.3 billion activated), 48 layers, 128 experts (8 activated per task), and supports up to 131K token contexts with YaRN, setting a new standard among open-source models.

Key Specifications
Cost
$$$$
Context
40K
Parameters
30B
Released
Apr 28, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Presence Penalty Temperature Seed Response Format Frequency Penalty Max Tokens Include Reasoning Tool Choice Top P Min P Tools Reasoning
Features

This model supports the following features:

Tools Reasoning Response Format
Performance Summary

Qwen3 30B A3B, the latest Qwen model, demonstrates exceptional performance across a range of benchmarks, particularly in its reliability. It achieves a perfect 100th percentile in reliability, consistently providing usable responses with minimal technical failures. While its speed tends to be slower, ranking in the 17th percentile, it offers competitive pricing, falling in the 49th percentile. The model excels in critical areas, achieving perfect 100% accuracy in both Ethics and General Knowledge, with the latter also being among the top three in accuracy and most accurate among models of comparable speed and price. It also shows strong capabilities in Reasoning (98% accuracy) and Coding (94% accuracy), placing it in the 95th percentile for both. Email Classification is also a strength at 99% accuracy. Its primary area for improvement is Instruction Following, where it achieved 60.8% accuracy. Overall, Qwen3 30B A3B stands out for its high accuracy in complex reasoning and knowledge-based tasks, coupled with outstanding reliability, making it a robust choice despite its longer response times.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.08
Completion $0.29

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
DeepInfra
DeepInfra | qwen/qwen3-30b-a3b-04-28 40K $0.08 / 1M tokens $0.29 / 1M tokens
InferenceNet
InferenceNet | qwen/qwen3-30b-a3b-04-28 16K $0.02 / 1M tokens $0.08 / 1M tokens
Parasail
Parasail | qwen/qwen3-30b-a3b-04-28 40K $0.09 / 1M tokens $0.5 / 1M tokens
Nebius
Nebius | qwen/qwen3-30b-a3b-04-28 40K $0.1 / 1M tokens $0.3 / 1M tokens
Novita
Novita | qwen/qwen3-30b-a3b-04-28 40K $0.1 / 1M tokens $0.45 / 1M tokens
Fireworks
Fireworks | qwen/qwen3-30b-a3b-04-28 131K $0.02 / 1M tokens $0.08 / 1M tokens
Friendli
Friendli | qwen/qwen3-30b-a3b-04-28 131K $0.15 / 1M tokens $0.6 / 1M tokens
NextBit
NextBit | qwen/qwen3-30b-a3b-04-28 32K $0.02 / 1M tokens $0.08 / 1M tokens
Chutes
Chutes | qwen/qwen3-30b-a3b-04-28 40K $0.02 / 1M tokens $0.08 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by qwen