Qwen: Qwen3 30B A3B Instruct 2507

Text input Text output
Author's Description

Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference. It operates in non-thinking mode and is designed for high-quality instruction following, multilingual understanding, and agentic tool use. Post-trained on instruction data, it demonstrates competitive performance across reasoning (AIME, ZebraLogic), coding (MultiPL-E, LiveCodeBench), and alignment (IFEval, WritingBench) benchmarks. It outperforms its non-instruct variant on subjective and open-ended tasks while retaining strong factual and coding performance.

Key Specifications
Cost
$$$
Context
131K
Parameters
30B
Released
Jul 29, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Response Format Seed Top P Max Tokens Temperature Presence Penalty
Features

This model supports the following features:

Response Format
Performance Summary

Qwen3 30B A3B Instruct 2507 demonstrates strong overall performance, particularly excelling in reliability with a 100% success rate across all benchmarks, indicating exceptional stability. The model performs among the fastest models, ranking in the 63rd percentile for speed, and offers competitive pricing, placing in the 56th percentile. Its strengths are evident in several key areas. It achieves perfect accuracy in both Keyword Topic Relevance Classification and Ethics, making it the most accurate model at its price point and speed for these tasks. The model also shows impressive capabilities in Mathematics (94.0% accuracy, 91st percentile) and Coding (93.0% accuracy, 88th percentile), suggesting robust reasoning and programming knowledge. Furthermore, its 98.0% accuracy in Hallucinations indicates a strong ability to acknowledge uncertainty. While its Instruction Following (59.8% accuracy) and Email Classification (94.0% accuracy, 30th percentile) are less stellar, they remain functional. The model's description highlights its design for high-quality instruction following and agentic tool use, which aligns with its strong performance in reasoning and coding.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.2
Completion $0.8

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Alibaba
Alibaba | qwen/qwen3-30b-a3b-instruct-2507 131K $0.2 / 1M tokens $0.8 / 1M tokens
Nebius
Nebius | qwen/qwen3-30b-a3b-instruct-2507 262K $0.1 / 1M tokens $0.3 / 1M tokens
Chutes
Chutes | qwen/qwen3-30b-a3b-instruct-2507 262K $0.07 / 1M tokens $0.28 / 1M tokens
SiliconFlow
SiliconFlow | qwen/qwen3-30b-a3b-instruct-2507 262K $0.09 / 1M tokens $0.3 / 1M tokens
AtlasCloud
AtlasCloud | qwen/qwen3-30b-a3b-instruct-2507 131K $0.09 / 1M tokens $0.45 / 1M tokens
Chutes
Chutes | qwen/qwen3-30b-a3b-instruct-2507 262K $0.07 / 1M tokens $0.28 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by qwen