Qwen: Qwen3 30B A3B Thinking 2507
QWEN Developer Architecture Profile
Intelligence (ELO)1430Chatbot Arena Verified
Max Context Limit81,920Max Out: 32,768 tokens
Prompt CachingStandardNo Cache Discount
Standard API / 1M$2.60🧠 Deep Reasoner
Model Capabilities & Contract SLAs
- Reasoning
- Agentic
- Coding & Logic
- Fictional
- 🛠️ Native Tool Calling
- 🧠 Extended Test-Time Reasoning
Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces are separated...
Granular Pricing Matrix
Input Tokens (Standard Prompt)$0.20 / 1M
Output Tokens (Completion)$2.40 / 1M
Pricing data via OpenRouter. Sync: 10/2/2026
Evaluate Competitors
VS Engine MatchupQwen: Qwen3 30B A3B Thinking 2507 vs Xiaomi: MiMo-V2.6-FlashVS Engine MatchupQwen: Qwen3 30B A3B Thinking 2507 vs StepFun: Step 3.7 FlashVS Engine MatchupQwen: Qwen3 30B A3B Thinking 2507 vs MiniMax: MiniMax M2.1VS Engine MatchupQwen: Qwen3 30B A3B Thinking 2507 vs MoonshotAI: Kimi K2 0711VS Engine MatchupQwen: Qwen3 30B A3B Thinking 2507 vs TheDrummer: UnslopNemo 12BVS Engine MatchupQwen: Qwen3 30B A3B Thinking 2507 vs inclusionAI: Ling 3.0 Flash VL