foundation modelinfrastructureapplications

Alibaba Cloud

Founded: 2009HQ: Hangzhou, China
Visit Websiteopen_in_new
Total FundingUndisclosed
ValuationUndisclosed
Headcount10000+
Founded Year2009

Profile Overview

Alibaba Cloud (阿里云智能) is the cloud computing and artificial intelligence division of Alibaba Group (阿里巴巴集团), one of the world's largest e-commerce and technology conglomerates. Alibaba Cloud operates as the AI infrastructure and foundation model arm of Alibaba, developing the Tongyi Qianwen (通义千问, abbreviated Qwen) family of large language models under the leadership of Dr. Wang Jian, the founder of Alibaba Cloud, and its AI research teams. Qwen (Qianwen) is Alibaba's flagship series of LLMs, spanning base models, instruction-tuned variants (Qwen2.5, Qwen2-VL, Qwen3), specialized code models (Qwen2.5-Coder), mathematics models (Qwen2.5-Math), and multimodal models that process text, images, audio, and video. Qwen models consistently rank among the world's top-performing open-weight models, frequently competing with Meta's Llama series on multilingual and coding benchmarks while offering superior Chinese-language capability. Alibaba Cloud also develops enterprise AI agents and applications, including Tongyi Lingma, an AI coding assistant, and Tongyi Wanxiang, a visual generation platform. Alibaba Cloud's AI services are integrated into Alibaba's e-commerce ecosystem, powering search, recommendations, and customer service across Taobao and Tmall. As a division of Alibaba Group ($27.5B annual cloud revenue), Alibaba Cloud does not raise external venture funding. Its AI research is supported by Alibaba's massive infrastructure, including proprietary Hanguang AI chips and one of Asia's largest GPU compute clusters.

Last Financing Round

Round TypeCorporate Funding
Amount DisclosedUndisclosed
Announced Date

Flagship Offerings

QwenTongyi LingmaModel Studio
SHARE COMPANY:

Vetted AI Models

Qwen3.8 Max (0902)

active

Qwen3.8 Max (0902) is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.8 Max (0902) represents a key architectural iteration in the Alibaba model family. First released in 2026-09-03, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.8 Max (0902) processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$2
Out Price$6

Qwen3.8 Flash

active

Qwen3.8 Flash is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.8 Flash represents a key architectural iteration in the Alibaba model family. First released in 2026-08-26, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.8 Flash processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.15
Out Price$0.47

Qwen3.8 27B

active

Qwen3.8 27B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.8 27B represents a key architectural iteration in the Alibaba model family. First released in 2026-08-14, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.8 27B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.214
Out Price$2.55

Qwen3.8 2.4T A95B

active

Qwen3.8 2.4T A95B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.8 2.4T A95B represents a key architectural iteration in the Alibaba model family. First released in 2026-08-12, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,048,576 tokens (approximately 1,398 words), Qwen3.8 2.4T A95B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1048.576k
In Price$2
Out Price$6

Qwen3.8 Max

active

Qwen3.8 Max is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.8 Max represents a key architectural iteration in the Alibaba model family. First released in 2026-08-03, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.8 Max processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$2
Out Price$6

Qwen3.7 Flash

active

Qwen3.7 Flash is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.7 Flash represents a key architectural iteration in the Alibaba model family. First released in 2026-07-27, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.7 Flash processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.03
Out Price$0.13

Qwen3.7 Plus

active

Qwen3.7 Plus is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.7 Plus represents a key architectural iteration in the Alibaba model family. First released in 2026-06-03, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.7 Plus processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.32
Out Price$1.28

Qwen3.7 Max

active

Qwen3.7 Max is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.7 Max represents a key architectural iteration in the Alibaba model family. First released in 2026-05-21, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.7 Max processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$1.475
Out Price$4.425

Qwen3.6 27B

active

Qwen3.6 27B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.6 27B represents a key architectural iteration in the Alibaba model family. First released in 2026-04-27, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.6 27B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.3
Out Price$2

Qwen3.6 Flash

active

Qwen3.6 Flash is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.6 Flash represents a key architectural iteration in the Alibaba model family. First released in 2026-04-27, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.6 Flash processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.1875
Out Price$1.125

Qwen3.5 Plus 2026-04-20

active

Qwen3.5 Plus 2026-04-20 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5 Plus 2026-04-20 represents a key architectural iteration in the Alibaba model family. First released in 2026-04-27, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.5 Plus 2026-04-20 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.3
Out Price$1.8

Qwen3.6 35B A3B

active

Qwen3.6 35B A3B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.6 35B A3B represents a key architectural iteration in the Alibaba model family. First released in 2026-04-27, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.6 35B A3B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.1
Out Price$0.9

Qwen3.6 Max Preview

active

Qwen3.6 Max Preview is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.6 Max Preview represents a key architectural iteration in the Alibaba model family. First released in 2026-04-27, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.6 Max Preview processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$1.027
Out Price$6.162

Qwen3.6 Plus

active

Qwen3.6 Plus is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.6 Plus represents a key architectural iteration in the Alibaba model family. First released in 2026-04-02, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.6 Plus processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.325
Out Price$1.95

Qwen3.5-9B

active

Qwen3.5-9B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5-9B represents a key architectural iteration in the Alibaba model family. First released in 2026-03-10, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.5-9B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.1
Out Price$0.15

Qwen3.5-Flash

active

Qwen3.5-Flash is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5-Flash represents a key architectural iteration in the Alibaba model family. First released in 2026-02-25, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.5-Flash processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.065
Out Price$0.26

Qwen3.5-35B-A3B

active

Qwen3.5-35B-A3B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5-35B-A3B represents a key architectural iteration in the Alibaba model family. First released in 2026-02-25, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.5-35B-A3B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.1625
Out Price$1.3

Qwen3.5-122B-A10B

active

Qwen3.5-122B-A10B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5-122B-A10B represents a key architectural iteration in the Alibaba model family. First released in 2026-02-25, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.5-122B-A10B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.26
Out Price$2.08

Qwen3.5-27B

active

Qwen3.5-27B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5-27B represents a key architectural iteration in the Alibaba model family. First released in 2026-02-25, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.5-27B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.195
Out Price$1.56

Qwen3.5 Plus 2026-02-15

active

Qwen3.5 Plus 2026-02-15 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5 Plus 2026-02-15 represents a key architectural iteration in the Alibaba model family. First released in 2026-02-16, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3.5 Plus 2026-02-15 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.26
Out Price$1.56

Qwen3.5 397B A17B

active

Qwen3.5 397B A17B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3.5 397B A17B represents a key architectural iteration in the Alibaba model family. First released in 2026-02-16, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3.5 397B A17B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.55
Out Price$3.5

Qwen3 Max Thinking

active

Qwen3 Max Thinking is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Max Thinking represents a key architectural iteration in the Alibaba model family. First released in 2026-02-09, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Max Thinking processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.78
Out Price$3.9

Qwen3 Coder Next

active

Qwen3 Coder Next is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Coder Next represents a key architectural iteration in the Alibaba model family. First released in 2026-02-04, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Coder Next processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.12
Out Price$0.8

Qwen 2.5-Coder 32B

active

Qwen 2.5-Coder 32B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen 2.5-Coder 32B represents a key architectural iteration in the Alibaba model family. First released in 2025-11-12, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen 2.5-Coder 32B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.35
Out Price$0.7

Qwen3 VL 32B Instruct

active

Qwen3 VL 32B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 32B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-10-23, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 VL 32B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.104
Out Price$0.416

Qwen3 VL 8B Instruct

active

Qwen3 VL 8B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 8B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-10-14, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 VL 8B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.117
Out Price$0.455

Qwen3 VL 8B Thinking

active

Qwen3 VL 8B Thinking is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 8B Thinking represents a key architectural iteration in the Alibaba model family. First released in 2025-10-14, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 VL 8B Thinking processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.18
Out Price$2.1

Qwen3 VL 30B A3B Thinking

active

Qwen3 VL 30B A3B Thinking is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 30B A3B Thinking represents a key architectural iteration in the Alibaba model family. First released in 2025-10-06, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 VL 30B A3B Thinking processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.2
Out Price$2.4

Qwen3 VL 30B A3B Instruct

active

Qwen3 VL 30B A3B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 30B A3B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-10-06, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 VL 30B A3B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.13
Out Price$0.52

Qwen3 VL 235B A22B Thinking

active

Qwen3 VL 235B A22B Thinking is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 235B A22B Thinking represents a key architectural iteration in the Alibaba model family. First released in 2025-09-23, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 VL 235B A22B Thinking processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.4
Out Price$4

Qwen3 VL 235B A22B Instruct

active

Qwen3 VL 235B A22B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 VL 235B A22B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-09-23, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 VL 235B A22B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.21
Out Price$1.9

Qwen3 Max

active

Qwen3 Max is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Max represents a key architectural iteration in the Alibaba model family. First released in 2025-09-23, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Max processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.78
Out Price$3.9

Qwen3 Coder Plus

active

Qwen3 Coder Plus is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Coder Plus represents a key architectural iteration in the Alibaba model family. First released in 2025-09-23, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3 Coder Plus processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.65
Out Price$3.25

Qwen 2.5 72B

active

Qwen 2.5 72B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen 2.5 72B represents a key architectural iteration in the Alibaba model family. First released in 2025-09-19, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen 2.5 72B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.4
Out Price$0.8

Qwen3 Coder Flash

active

Qwen3 Coder Flash is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Coder Flash represents a key architectural iteration in the Alibaba model family. First released in 2025-09-17, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen3 Coder Flash processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.195
Out Price$0.975

Qwen3 Next 80B A3B Thinking

active

Qwen3 Next 80B A3B Thinking is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Next 80B A3B Thinking represents a key architectural iteration in the Alibaba model family. First released in 2025-09-11, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Next 80B A3B Thinking processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.15
Out Price$1.2

Qwen3 Next 80B A3B Instruct

active

Qwen3 Next 80B A3B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Next 80B A3B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-09-11, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Next 80B A3B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.09
Out Price$1.1

Qwen Plus 0728 (thinking)

active

Qwen Plus 0728 (thinking) is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen Plus 0728 (thinking) represents a key architectural iteration in the Alibaba model family. First released in 2025-09-08, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen Plus 0728 (thinking) processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.26
Out Price$0.78

Qwen Plus 0728

active

Qwen Plus 0728 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen Plus 0728 represents a key architectural iteration in the Alibaba model family. First released in 2025-09-08, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen Plus 0728 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.26
Out Price$0.78

Qwen3 30B A3B Thinking 2507

active

Qwen3 30B A3B Thinking 2507 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 30B A3B Thinking 2507 represents a key architectural iteration in the Alibaba model family. First released in 2025-08-28, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 81,920 tokens (approximately 109 words), Qwen3 30B A3B Thinking 2507 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context81.92k
In Price$0.2
Out Price$2.4

Qwen3 Coder 30B A3B Instruct

active

Qwen3 Coder 30B A3B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Coder 30B A3B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-07-31, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Coder 30B A3B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.07
Out Price$0.28

Qwen3 30B A3B Instruct 2507

active

Qwen3 30B A3B Instruct 2507 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 30B A3B Instruct 2507 represents a key architectural iteration in the Alibaba model family. First released in 2025-07-29, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 30B A3B Instruct 2507 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.0481
Out Price$0.193

Qwen3 235B A22B Thinking 2507

active

Qwen3 235B A22B Thinking 2507 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 235B A22B Thinking 2507 represents a key architectural iteration in the Alibaba model family. First released in 2025-07-25, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 235B A22B Thinking 2507 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.23
Out Price$2.3

Qwen3 Coder 480B A35B

active

Qwen3 Coder 480B A35B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 Coder 480B A35B represents a key architectural iteration in the Alibaba model family. First released in 2025-07-23, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 Coder 480B A35B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.3
Out Price$1

Qwen3 235B A22B Instruct 2507

active

Qwen3 235B A22B Instruct 2507 is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 235B A22B Instruct 2507 represents a key architectural iteration in the Alibaba model family. First released in 2025-07-21, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 262,144 tokens (approximately 350 words), Qwen3 235B A22B Instruct 2507 processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context262.144k
In Price$0.0875
Out Price$0.35

Qwen3 14B

active

Qwen3 14B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 14B represents a key architectural iteration in the Alibaba model family. First released in 2025-04-28, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 14B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.12
Out Price$0.24

Qwen3 30B A3B

active

Qwen3 30B A3B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 30B A3B represents a key architectural iteration in the Alibaba model family. First released in 2025-04-28, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 30B A3B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.12
Out Price$0.5

Qwen3 8B

active

Qwen3 8B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 8B represents a key architectural iteration in the Alibaba model family. First released in 2025-04-28, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 8B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.117
Out Price$0.455

Qwen3 32B

active

Qwen3 32B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 32B represents a key architectural iteration in the Alibaba model family. First released in 2025-04-28, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 32B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.08
Out Price$0.28

Qwen3 235B A22B

active

Qwen3 235B A22B is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen3 235B A22B represents a key architectural iteration in the Alibaba model family. First released in 2025-04-28, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 131,072 tokens (approximately 175 words), Qwen3 235B A22B processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context131.072k
In Price$0.455
Out Price$1.82

Qwen2.5 VL 72B Instruct

active

Qwen2.5 VL 72B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen2.5 VL 72B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2025-02-01, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 128,000 tokens (approximately 171 words), Qwen2.5 VL 72B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context128k
In Price$0.8
Out Price$1

Qwen-Plus

active

Qwen-Plus is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen-Plus represents a key architectural iteration in the Alibaba model family. First released in 2025-02-01, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 1,000,000 tokens (approximately 1,333 words), Qwen-Plus processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context1000k
In Price$0.26
Out Price$0.78

Qwen2.5 Coder 32B Instruct

active

Qwen2.5 Coder 32B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen2.5 Coder 32B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2024-11-11, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 32,768 tokens (approximately 44 words), Qwen2.5 Coder 32B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context32.768k
In Price$0.66
Out Price$1

Qwen2.5 7B Instruct

active

Qwen2.5 7B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen2.5 7B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2024-10-16, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 32,768 tokens (approximately 44 words), Qwen2.5 7B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context32.768k
In Price$0.1
Out Price$0.2

Qwen2.5 72B Instruct

active

Qwen2.5 72B Instruct is an advanced artificial intelligence model engineered by Alibaba. Tailored for complex multi-step reasoning, programming synthesis, and extended document comprehension, Qwen2.5 72B Instruct represents a key architectural iteration in the Alibaba model family. First released in 2024-09-19, it serves enterprise developers, research teams, and autonomous system architects requiring strict instruction compliance. Featuring an input capacity of 32,768 tokens (approximately 44 words), Qwen2.5 72B Instruct processes multi-file code repositories, lengthy technical reports, and complex prompts in a single inference call.

Context32.768k
In Price$0.36
Out Price$0.4