Model
Qwen2
Qwen2 is the Qwen family's June 2024 generation, with five sizes from 0.5B to 72B parameters, context lengths up to 128K tokens, and most sizes released under Apache 2.0.
Status DeprecatedModality TextOpen weights YesContext (K tokens) 128 Era Generative AI for everyone
- Parameters
- 72B
- License
- Apache-2.0
- Access
- API, Download
- Release date
- Developer
- Alibaba Cloud
What it is
Qwen2 is the second major generation of the Qwen large language model family from Alibaba, released on Hugging Face and ModelScope. It comes as pretrained base models and instruction-tuned models, so each size is available in both forms. The five sizes are 0.5B, 1.5B, 7B, 57B-A14B and 72B.[1]
The 57B-A14B model is a mixture-of-experts design. It has 57 billion parameters in total, of which 14 billion are active for each input, because the model routes each input to a subset of its expert sub-networks.[1]
Origins
The Qwen repository dates the release to 6 June 2024, and the Qwen team's blog post is dated 7 June 2024.[2][1] Qwen2 succeeded Qwen1.5.[2] It was followed by Qwen2.5 on 19 September 2024.[2]
Main features
The context length depends on the size. The 0.5B and 1.5B models support 32K tokens, the 57B-A14B supports 64K, and the 7B and 72B support 128K. The instruct models reach 128K with YaRN, a technique that extends the length a model can handle.[1]
The training data covered 27 languages beyond English and Chinese.[1] Qwen2 uses Grouped Query Attention at all sizes. This is a way of organising the attention layers so that groups of query heads share key and value heads, which reduces memory use during generation. The two smallest models also use tied embeddings, which means the input and output word-embedding matrices share the same weights.[1]
Licensing
The licences differed by size. Qwen2-72B kept the Qianwen License, while the 0.5B, 1.5B, 7B and 57B-A14B models moved to Apache 2.0.[1] Apache 2.0 is a permissive open-source licence that allows commercial use, with conditions such as keeping copyright notices.
Results
The Qwen team compared Qwen2-72B with Llama-3-70B. On the base models, Qwen2-72B scored 84.2 against 79.5 on MMLU, 64.6 against 48.2 on HumanEval, 89.5 against 83.0 on GSM8K, and 51.1 against 42.5 on MATH.[1]
For the instruction-tuned versions, Qwen2-72B-Instruct scored 86.0 against 81.7 on HumanEval, 59.7 against 50.4 on MATH, 9.12 against 8.95 on MT-Bench, and 48.1 against 41.1 on Arena-Hard.[1] These are the team's reported figures.
Its place in development
Qwen2 was followed in August 2024 by Qwen2-Audio, Qwen2-VL and Qwen2-Math, which extended the family to audio, vision and mathematics.[3] Qwen2 was in turn succeeded by Qwen2.5 on 19 September 2024.[2]
Version history
14-
2023
-
Aug 3, 2023
Qwen (Tongyi Qianwen, 2023)
Qwen, known in Chinese as Tongyi Qianwen (通义千问), is Alibaba's family of large language models. Its first open-weight model, Qwen-7B, was released on 3 August 2023, and the family has grown through many later versions.
-
Aug 3, 2023
Qwen (Tongyi Qianwen, 2023)
-
2024
-
Feb 5, 2024
Qwen1.5
Qwen1.5 is the Qwen family's open model release of February 2024, with base and chat models from 0.5B to 72B parameters, later extended to 110B, and a mixture-of-experts model.
-
Jun 7, 2024
Qwen2 · This page
Qwen2 is the Qwen family's June 2024 generation, with five sizes from 0.5B to 72B parameters, context lengths up to 128K tokens, and most sizes released under Apache 2.0.
-
Sep 19, 2024
Qwen2.5
Qwen2.5 is the Qwen family's September 2024 generation. Its launch included 13 open-weight models, with up to 18 trillion pretraining tokens, 128K context and most models under Apache 2.0.
-
Nov 28, 2024
QwQ-32B-Preview
QwQ-32B-Preview is an experimental reasoning research model released by the Qwen team on 28 November 2024, with 32B parameters, a 32K-token context and open Apache 2.0 weights. It scored strongly on maths benchmarks but the team documented clear failure modes.
-
Feb 5, 2024
Qwen1.5
-
2025
-
Jan 28, 2025
Qwen2.5-Max
Qwen2.5-Max is Alibaba's large mixture-of-experts model, announced in January 2025. It was offered through Qwen Chat and the Alibaba Cloud API, and its weights were not released.
-
Mar 6, 2025
QwQ-32B
A 32-billion-parameter open-weight reasoning model released by the Qwen team on 6 March 2025, trained mainly through scaled reinforcement learning and reported to perform comparably to the much larger DeepSeek-R1.
-
Apr 29, 2025
Qwen3
Qwen3 is a family of large language models from Alibaba, released on 29 April 2025 under Apache 2.0. Its main feature is hybrid thinking, which lets one model reason step by step or answer directly.
-
Jul 21, 2025
Qwen3-2507
Qwen3-2507 is a staged 2025 refresh of the open-weight Qwen3 models that split thinking and non-thinking into separate checkpoints, raised native context to 262,144 tokens, and scored far above the original non-thinking mode on published benchmarks.
-
Sep 2025
Qwen3-Max
Qwen3-Max is Alibaba's proprietary trillion-parameter language model, released in September 2025 and served through Alibaba's API. It is the Max-tier flagship that followed Qwen2.5-Max.
-
Sep 2025
Qwen3-Next
Qwen3-Next is an open-weight model from Alibaba's Qwen team, released in September 2025 under Apache 2.0. It has 80B total parameters, only 3B active per token, and a hybrid attention design later used in Qwen3.5.
-
Jan 28, 2025
Qwen2.5-Max
-
2026
-
Feb 16, 2026
Qwen3.5
Qwen3.5 is a natively multimodal model family from Alibaba released in February 2026. Its open flagship, Qwen3.5-397B-A17B, has 397B total parameters and 17B active per token, under Apache 2.0.
-
Apr 2, 2026
Qwen3.6
A generation of Alibaba's Qwen models released in April 2026, spanning proprietary Plus and Max versions aimed at agentic coding and open-weight models such as Qwen3.6-35B-A3B, a 35B-parameter mixture-of-experts with 3B active parameters.
-
Aug 3, 2026
Qwen3.8
Qwen3.8 is Alibaba's 2026 flagship line. Its open 2.4-trillion-parameter model, Qwen3.8-2.4T-A95B, released in August 2026, has 95B active parameters and a custom license.
-
Feb 16, 2026
Qwen3.5
20241
-
ReleaseAlibaba releases Qwen2, moving most sizes to Apache 2.0
On 6 and 7 June 2024 Alibaba released Qwen2 on Hugging Face and ModelScope. Most sizes moved to Apache 2.0, and the 72B model kept the Qianwen License.
Developed by1
Foundation for1
Succeeded by1
Successor of1
Notes and sources
- Hello Qwen2https://qwenlm.github.io/blog/qwen2/
- QwenLM/Qwen3 repositoryhttps://github.com/QwenLM/Qwen3
- Qwenhttps://en.wikipedia.org/wiki/Qwen
Cite this page: Qwen2. TellWorld, https://www.tellworld.net/models/qwen2. Updated 2026-11-08