Model Family · Alibaba

Qwen

Name: Qwen model family
Creator: Alibaba

Qwen is Alibaba’s open-weight model family. The Qwen series leads many open leaderboards for code, math, and multilingual benchmarks and includes Mixture-of-Experts and dense variants from 0.5B to 235B parameters.

See all models from Alibaba

Models in family

Open weight

API only

Avg score

65.1

Top benchmark

95.3

Arena Score

Total HF downloads

41.6M

Primary modality

Text

Context window

131K – 1.0M

First release

Jan 2025

Latest release

Apr 2026

Qwen Family Leaderboard

Every release in the Qwen family, ranked by composite score across benchmarks, popularity, efficiency, and versatility.

#	Model	Modality	Score	Params	Released
1	Qwen3-ASR-0.6B	audio	AA83.4	—	Jan 2026
2	Qwen3.6 Max PreviewAPI	text	AA77.7	—	Apr 2026
3	Qwen3.5-9B	text	AA77.5	9B	Mar 2026
4	Qwen3.5-35B-A3B	text	AA77.1	35B	Feb 2026
5	Qwen3-Embedding-0.6B	embedding	AA74.7	—	Jun 2025
6	Qwen3.5 FlashAPI	text	AA73.9	35B	Feb 2026
7	Qwen3-Embedding-4B	embedding	AA72.2	4B	Jun 2025
8	Qwen3.6 35B-A3B	text	AA72.0	35B	Apr 2026
9	Qwen3-Embedding-8B	embedding	AA71.9	7.6B	Jun 2025
10	Qwen3.5 Max PreviewAPI	text	AA71.9	—	Feb 2026

See All Models

Best in Each Modality

When a family spans multiple modalities, the strongest model in each one.

Text

Qwen3.6 Max Preview

77.7

Image

Qwen-Image-2512

47.4

Audio

Qwen3-ASR-0.6B

83.4

Embedding

Qwen3-Embedding-0.6B

74.7

Modality Breakdown

How the family splits across text, image, video, audio, and embedding releases.

Text19 (68%)

Image4 (14%)

Audio2 (7%)

Embedding3 (11%)

Release Timeline

When each release shipped, newest first. Useful for tracking version cadence.

2026
- Qwen3.6-27B
  Apr 21
  65.2
- Qwen3.6 Max Preview
  Apr 17
  77.7
- Qwen3.6 35B-A3B
  Apr 13
  72.0
- Qwen3.6-Plus
  Mar 31
  52.5
- Qwen 3.5 Omni
  Mar 28
  64.2
- Qwen3.5-9B
  Mar 1
  77.5
- Qwen3.5-122B-A10B
  Feb 23
  62.0
- Qwen3.5-35B-A3B
  Feb 23
  77.1
- Qwen3.5-27B
  Feb 23
  64.5
- Qwen3.5 Flash
  Feb 23
  73.9
- Qwen3.5-397B-A17B
  Feb 15
  71.1
- Qwen3.5 Max Preview
  Feb 15
  71.9
- Qwen3-ASR-1.7B
  Jan 29
  71.8
- Qwen3-ASR-0.6B
  Jan 29
  83.4
2025
- Qwen-Image-2512
  Dec 1
  47.4
- Qwen-Image-Edit-2511
  Nov 1
  45.2
- Qwen3 Max Preview
  Sep 22
  66.9
- Qwen3 Max (2025-09-23)
  Sep 22
  64.9
- Qwen3-Embedding-8B
  Jun 5
  71.9
- Qwen3-Embedding-4B
  Jun 5
  72.2
- Qwen3-Embedding-0.6B
  Jun 5
  74.7
- Qwen3-235B-A22B
  Apr 27
  54.8
- Qwen3-32B
  Apr 27
  67.9
- Qwen3-30B-A3B
  Apr 27
  70.2
- Qwen2.5 Max
  Jan 24
  61.2
- Qwen Plus (0125)
  Jan 24
  60.2

Grade Distribution

Composite grades across this family. Higher is better, blending benchmarks, popularity, and efficiency.

AGrade A13 (46%)

BGrade B9 (32%)

CGrade C5 (18%)

DGrade D1 (4%)

FGrade F0 (0%)

Open-Weight Qwen Releases

Models with downloadable weights, ranked by composite score.

#	Model	Modality	Score	Params	Released
1	Qwen3-ASR-0.6B	audio	AA83.4	—	Jan 2026
2	Qwen3.5-9B	text	AA77.5	9B	Mar 2026
3	Qwen3.5-35B-A3B	text	AA77.1	35B	Feb 2026
4	Qwen3-Embedding-0.6B	embedding	AA74.7	—	Jun 2025
5	Qwen3-Embedding-4B	embedding	AA72.2	4B	Jun 2025
6	Qwen3.6 35B-A3B	text	AA72.0	35B	Apr 2026
7	Qwen3-Embedding-8B	embedding	AA71.9	7.6B	Jun 2025
8	Qwen3-ASR-1.7B	audio	AA71.8	1.7B	Jan 2026
9	Qwen3.5-397B-A17B	text	AA71.1	397B	Feb 2026
10	Qwen3-30B-A3B	text	AA70.2	30B	Apr 2025
11	Qwen3-32B	text	BB67.9	32.8B	Apr 2025
12	Qwen3.6-27B	text	BB65.2	27B	Apr 2026
13	Qwen3.5-27B	text	BB64.5	27B	Feb 2026
14	Qwen 3.5 Omni	text	BB64.2	397B	Mar 2026
15	Qwen3.5-122B-A10B	text	BB62.0	122B	Feb 2026
16	Qwen3-235B-A22B	text	CC54.8	235B	Apr 2025
17	Qwen-Image-2512	image	CC47.4	20B	Dec 2025
18	Qwen-Image-Edit	image	CC46.8	20B	—
19	Qwen-Image-Edit-2511	image	CC45.2	20B	Nov 2025
20	Qwen-Image	image	DD32.8	20B	—

API-Only Qwen Releases

Closed-source releases accessed through Alibaba’s API.

#	Model	Modality	Score	Params	Released
1	Qwen3.6 Max PreviewAPI	text	AA77.7	—	Apr 2026
2	Qwen3.5 FlashAPI	text	AA73.9	35B	Feb 2026
3	Qwen3.5 Max PreviewAPI	text	AA71.9	—	Feb 2026
4	Qwen3 Max PreviewAPI	text	BB66.9	—	Sep 2025
5	Qwen3 Max (2025-09-23)API	text	BB64.9	—	Sep 2025
6	Qwen2.5 MaxAPI	text	BB61.2	—	Jan 2025
7	Qwen Plus (0125)API	text	BB60.2	—	Jan 2025
8	Qwen3.6-PlusAPI	text	CC52.5	—	Mar 2026

Sibling Families From the Same Provider

Other model series from this lab worth comparing.

Wan

CosyVoice

Frequently Asked Questions

What is the Qwen family?

The Qwen family is a series of AI models from Alibaba. This page lists every release in the family with its benchmark scores, parameter count, and hardware requirements.

Which Qwen model should I use?

By composite score, Qwen3-ASR-0.6B is currently the top model in the family. For local inference, match the parameter count to your VRAM budget. For quality, pick the highest scorer that fits.

Are Qwen models open source?

See the open-weight section above for models you can run locally. The API-only section lists closed releases that must be accessed through the provider’s API.

Where to Run Qwen?

Spin up an instance in the cloud, or pick local hardware that fits.

Full Directory

Compare With Every Other Model

Open the full directory to filter by hardware, capability, license, and benchmark score.

Open Directory

Or Browse by Provider

Look up OpenAI, Anthropic, Meta, and Every Provider

See every model from a lab side by side, with aggregate stats. Useful when you want a cross-family view of one provider.

Browse Providers

Need Help Picking a Family Member?

We help teams match a model size to their hardware, budget, and latency target. Free 20-minute call.

Model Family · Alibaba

Qwen

See all models from Alibaba

Models in family

Open weight

API only

Avg score

65.1

Top benchmark

95.3

Arena Score

Total HF downloads

41.6M

Primary modality

Text

Context window

131K – 1.0M

First release

Jan 2025

Latest release

Apr 2026

Qwen Family Leaderboard

Every release in the Qwen family, ranked by composite score across benchmarks, popularity, efficiency, and versatility.

#	Model	Modality	Score	Params	Released
1	Qwen3-ASR-0.6B	audio	AA83.4	—	Jan 2026
2	Qwen3.6 Max PreviewAPI	text	AA77.7	—	Apr 2026
3	Qwen3.5-9B	text	AA77.5	9B	Mar 2026
4	Qwen3.5-35B-A3B	text	AA77.1	35B	Feb 2026
5	Qwen3-Embedding-0.6B	embedding	AA74.7	—	Jun 2025
6	Qwen3.5 FlashAPI	text	AA73.9	35B	Feb 2026
7	Qwen3-Embedding-4B	embedding	AA72.2	4B	Jun 2025
8	Qwen3.6 35B-A3B	text	AA72.0	35B	Apr 2026
9	Qwen3-Embedding-8B	embedding	AA71.9	7.6B	Jun 2025
10	Qwen3.5 Max PreviewAPI	text	AA71.9	—	Feb 2026

See All Models

Best in Each Modality

When a family spans multiple modalities, the strongest model in each one.

Text

Qwen3.6 Max Preview

77.7

Image

Qwen-Image-2512

47.4

Audio

Qwen3-ASR-0.6B

83.4

Embedding

Qwen3-Embedding-0.6B

74.7

Modality Breakdown

How the family splits across text, image, video, audio, and embedding releases.

Text19 (68%)

Image4 (14%)

Audio2 (7%)

Embedding3 (11%)

Release Timeline

When each release shipped, newest first. Useful for tracking version cadence.

2026
- Qwen3.6-27B
  Apr 21
  65.2
- Qwen3.6 Max Preview
  Apr 17
  77.7
- Qwen3.6 35B-A3B
  Apr 13
  72.0
- Qwen3.6-Plus
  Mar 31
  52.5
- Qwen 3.5 Omni
  Mar 28
  64.2
- Qwen3.5-9B
  Mar 1
  77.5
- Qwen3.5-122B-A10B
  Feb 23
  62.0
- Qwen3.5-35B-A3B
  Feb 23
  77.1
- Qwen3.5-27B
  Feb 23
  64.5
- Qwen3.5 Flash
  Feb 23
  73.9
- Qwen3.5-397B-A17B
  Feb 15
  71.1
- Qwen3.5 Max Preview
  Feb 15
  71.9
- Qwen3-ASR-1.7B
  Jan 29
  71.8
- Qwen3-ASR-0.6B
  Jan 29
  83.4
2025
- Qwen-Image-2512
  Dec 1
  47.4
- Qwen-Image-Edit-2511
  Nov 1
  45.2
- Qwen3 Max Preview
  Sep 22
  66.9
- Qwen3 Max (2025-09-23)
  Sep 22
  64.9
- Qwen3-Embedding-8B
  Jun 5
  71.9
- Qwen3-Embedding-4B
  Jun 5
  72.2
- Qwen3-Embedding-0.6B
  Jun 5
  74.7
- Qwen3-235B-A22B
  Apr 27
  54.8
- Qwen3-32B
  Apr 27
  67.9
- Qwen3-30B-A3B
  Apr 27
  70.2
- Qwen2.5 Max
  Jan 24
  61.2
- Qwen Plus (0125)
  Jan 24
  60.2

Grade Distribution

Composite grades across this family. Higher is better, blending benchmarks, popularity, and efficiency.

AGrade A13 (46%)

BGrade B9 (32%)

CGrade C5 (18%)

DGrade D1 (4%)

FGrade F0 (0%)

Open-Weight Qwen Releases

Models with downloadable weights, ranked by composite score.

#	Model	Modality	Score	Params	Released
1	Qwen3-ASR-0.6B	audio	AA83.4	—	Jan 2026
2	Qwen3.5-9B	text	AA77.5	9B	Mar 2026
3	Qwen3.5-35B-A3B	text	AA77.1	35B	Feb 2026
4	Qwen3-Embedding-0.6B	embedding	AA74.7	—	Jun 2025
5	Qwen3-Embedding-4B	embedding	AA72.2	4B	Jun 2025
6	Qwen3.6 35B-A3B	text	AA72.0	35B	Apr 2026
7	Qwen3-Embedding-8B	embedding	AA71.9	7.6B	Jun 2025
8	Qwen3-ASR-1.7B	audio	AA71.8	1.7B	Jan 2026
9	Qwen3.5-397B-A17B	text	AA71.1	397B	Feb 2026
10	Qwen3-30B-A3B	text	AA70.2	30B	Apr 2025
11	Qwen3-32B	text	BB67.9	32.8B	Apr 2025
12	Qwen3.6-27B	text	BB65.2	27B	Apr 2026
13	Qwen3.5-27B	text	BB64.5	27B	Feb 2026
14	Qwen 3.5 Omni	text	BB64.2	397B	Mar 2026
15	Qwen3.5-122B-A10B	text	BB62.0	122B	Feb 2026
16	Qwen3-235B-A22B	text	CC54.8	235B	Apr 2025
17	Qwen-Image-2512	image	CC47.4	20B	Dec 2025
18	Qwen-Image-Edit	image	CC46.8	20B	—
19	Qwen-Image-Edit-2511	image	CC45.2	20B	Nov 2025
20	Qwen-Image	image	DD32.8	20B	—

API-Only Qwen Releases

Closed-source releases accessed through Alibaba’s API.

#	Model	Modality	Score	Params	Released
1	Qwen3.6 Max PreviewAPI	text	AA77.7	—	Apr 2026
2	Qwen3.5 FlashAPI	text	AA73.9	35B	Feb 2026
3	Qwen3.5 Max PreviewAPI	text	AA71.9	—	Feb 2026
4	Qwen3 Max PreviewAPI	text	BB66.9	—	Sep 2025
5	Qwen3 Max (2025-09-23)API	text	BB64.9	—	Sep 2025
6	Qwen2.5 MaxAPI	text	BB61.2	—	Jan 2025
7	Qwen Plus (0125)API	text	BB60.2	—	Jan 2025
8	Qwen3.6-PlusAPI	text	CC52.5	—	Mar 2026

Sibling Families From the Same Provider

Other model series from this lab worth comparing.

Wan

CosyVoice

Frequently Asked Questions

What is the Qwen family?

The Qwen family is a series of AI models from Alibaba. This page lists every release in the family with its benchmark scores, parameter count, and hardware requirements.

Which Qwen model should I use?

By composite score, Qwen3-ASR-0.6B is currently the top model in the family. For local inference, match the parameter count to your VRAM budget. For quality, pick the highest scorer that fits.

Are Qwen models open source?

See the open-weight section above for models you can run locally. The API-only section lists closed releases that must be accessed through the provider’s API.

Where to Run Qwen?

Spin up an instance in the cloud, or pick local hardware that fits.

Full Directory

Compare With Every Other Model

Open the full directory to filter by hardware, capability, license, and benchmark score.

Open Directory

Or Browse by Provider

Look up OpenAI, Anthropic, Meta, and Every Provider

See every model from a lab side by side, with aggregate stats. Useful when you want a cross-family view of one provider.

Browse Providers

Need Help Picking a Family Member?

We help teams match a model size to their hardware, budget, and latency target. Free 20-minute call.

Qwen

Quick Facts

Qwen Family Leaderboard

Best in Each Modality

Modality Breakdown

Release Timeline

2026

2025

Grade Distribution

Open-Weight Qwen Releases

API-Only Qwen Releases

Sibling Families From the Same Provider

Frequently Asked Questions

What is the Qwen family?

Which Qwen model should I use?

Are Qwen models open source?

Where to Run Qwen?

Compare With Every Other Model

Look up OpenAI, Anthropic, Meta, and Every Provider

Need Help Picking a Family Member?

Qwen

Quick Facts

Qwen Family Leaderboard

Best in Each Modality

Modality Breakdown

Release Timeline

2026

2025

Grade Distribution

Open-Weight Qwen Releases

API-Only Qwen Releases

Sibling Families From the Same Provider

Frequently Asked Questions

What is the Qwen family?

Which Qwen model should I use?

Are Qwen models open source?

Where to Run Qwen?

Compare With Every Other Model

Look up OpenAI, Anthropic, Meta, and Every Provider

Need Help Picking a Family Member?