Project AURIONAI Use-Cases Research & Innovation
Alibaba (Qwen)

Alibaba (Qwen)

Big tech, China. 155 models in the directory.

big techChina
0deployments

Identity

Type
big tech
Country label
China
HQ country
China
Founding country
China
Parent
not in the notes
Parent country
not in the notes
Country basis
the research notes
Units
Alibaba DAMOAlibaba DAMOFunAudioLLMFunAudioLLM
Where this comes from
Convention D1: a dual label, HQ first, when the HQ, founding or parent country differ.

Models by Class

LLM42

Models

ModelClassOpennessParametersReleased
Alibaba (Qwen)Qwen-Image-2.1image generationopen weights restricted7B20 Sep 2026Alibaba (Qwen)Qwen-Audio-3.1-TTS-Plusspeech synthesisunknown–Sep 2026Alibaba (Qwen)Qwen3.8-Flash-Nextmultimodal LLMopen weights restricted125B24 Aug 2026Alibaba (Qwen)Qwen3.8LLMunknown–8 Aug 2026Alibaba (Qwen)Wan 3.0video generationproprietary–6 Aug 2026Alibaba (Qwen)Qwen3.8 Maxmultimodal LLMproprietary–3 Aug 2026Alibaba (Qwen)Qwen3.8-2.4T-A95Breasoningopen weights restricted2.4TAug 2026Alibaba (Qwen)Qwen3.8-27Bmultimodal LLMopen weights permissive27BAug 2026Alibaba (Qwen)Wan2.2-Animate-2-14Bavataropen weights permissive14BAug 2026Alibaba (Qwen)Qwen-Image-3.0image generationproprietary–Jul 2026Alibaba (Qwen)Qwen-Image-3.0-Proimage generationproprietary–Jul 2026Alibaba (Qwen)Wan-Dancer-14Bavataropen weights permissive14BJul 2026Alibaba (Qwen)Qwen-AgentWorld-35B-A3BLLMopen weights permissive34.66B23 Jun 2026Alibaba (Qwen)HappyHorse 1.1video generationproprietary–Jun 2026Alibaba (Qwen)Qwen3.6LLMopen weights permissive–21 Apr 2026Alibaba (Qwen)Happy Oysterworld modelproprietary–16 Apr 2026Alibaba (Qwen)Qwen3.6-35B-A3BLLMopen weights permissive35B2 Apr 2026Alibaba (Qwen)HappyHorse 1.0video generationproprietary–Apr 2026Alibaba (Qwen)Qwen3.6-27Bmultimodal LLMopen weights permissive27BApr 2026Alibaba (Qwen)Qwen3.6 Max PreviewLLMproprietary1TApr 2026Alibaba (Qwen)Qwen 3.6 PlusLLMproprietary–Apr 2026Alibaba (Qwen)Qwen3-VL-Embedding-2Bembeddingopen weights permissive2BApr 2026Alibaba (Qwen)Qwen3-VL-Embedding-8Bembeddingopen weights permissive8BApr 2026Alibaba (Qwen)Qwen3.5LLMopen weights permissive–2 Mar 2026Alibaba (Qwen)Qwen3.5-0.8Bmultimodal LLMopen weights permissive800M2 Mar 2026Alibaba (Qwen)Qwen3.5-2Bmultimodal LLMopen weights permissive2B2 Mar 2026Alibaba (Qwen)Qwen3.5-35B-A3BLLMopen weights permissive35B2 Mar 2026Alibaba (Qwen)Qwen3.5-4Bmultimodal LLMopen weights permissive4B2 Mar 2026Alibaba (Qwen)Qwen3.5-9Bmultimodal LLMopen weights permissive9B2 Mar 2026Alibaba (Qwen)Qwen3.5 small (0.8B, 2B, 4B, 9B)multimodal LLMopen weights permissive–2 Mar 2026Alibaba (Qwen)Qwen3-Coder-Nextcodeopen weights permissive80B3 Feb 2026Alibaba (Qwen)Qwen3.5-122B-A10Bmultimodal LLMopen weights permissive122BFeb 2026Alibaba (Qwen)Qwen3.5-397B-A17Bmultimodal LLMopen weights permissive397BFeb 2026Alibaba (Qwen)Qwen3-ASRspeech recognitionopen weights permissive–29 Jan 2026Alibaba (Qwen)Qwen3-ASR-0.6Bspeech recognitionopen weights permissive600M29 Jan 2026Alibaba (Qwen)Qwen3-ASR-1.7Bspeech recognitionopen weights permissive1.7B29 Jan 2026Alibaba (Qwen)Qwen3-TTS 12Hz 0.6Bspeech synthesisopen weights permissive600M22 Jan 2026Alibaba (Qwen)Qwen3-TTS 12Hz 1.7Bspeech synthesisopen weights permissive1.7B22 Jan 2026Alibaba (Qwen)Qwen3-VL-Rerankerrerankeropen weights permissive–7 Jan 2026Alibaba (Qwen)Qwen3-VL-Reranker-2Brerankeropen weights permissive2.13B7 Jan 2026Alibaba (Qwen)Qwen3-VL-Reranker-8Brerankeropen weights permissive8.77B7 Jan 2026Alibaba (Qwen)Qwen-Image-2512image generationunknown–30 Dec 2025Alibaba (Qwen)Fun-ASR-Nano (Fun-ASR-Nano-2512)speech recognitionopen weights permissive800M15 Dec 2025Alibaba (Qwen)Qwen-Image-Layeredimage generationunknown–Dec 2025Alibaba (Qwen)Ovis-Image 7Bimage generationopen weights permissive7.37B28 Nov 2025Alibaba (Qwen)Z-Image Turboimage generationopen weights permissive6B27 Nov 2025Alibaba (Qwen)Qwen3Guardguardopen weights permissive–16 Oct 2025Alibaba (Qwen)Qwen3Guard-Gen-0.6Bguardopen weights permissive600M16 Oct 2025Alibaba (Qwen)Qwen3Guard-Gen-4Bguardopen weights permissive4B16 Oct 2025Alibaba (Qwen)Qwen3Guard-Gen-8Bguardopen weights permissive8B16 Oct 2025Alibaba (Qwen)Qwen3Guard-Stream-0.6Bguardopen weights permissive600M16 Oct 2025Alibaba (Qwen)Qwen3Guard-Stream-4Bguardopen weights permissive4B16 Oct 2025Alibaba (Qwen)Qwen3Guard-Stream-8Bguardopen weights permissive8B16 Oct 2025Alibaba (Qwen)Qwen3-VL-8B-Instructmultimodal LLMunknown8B11 Oct 2025Alibaba (Qwen)Qwen3-VL-235B-A22B-Instructmultimodal LLMopen weights permissive235.67B22 Sep 2025Alibaba (Qwen)Qwen3-Omni-30B-A3B-Instructmultimodal LLMopen weights permissive35.26B20 Sep 2025Alibaba (Qwen)Qwen3-Next-80B-A3B-InstructLLMopen weights permissive81.32B9 Sep 2025Alibaba (Qwen)Wan2.2-S2V-14Bvideo generationopen weights permissive16.3B25 Aug 2025Alibaba (Qwen)Qwen-Imageimage generationunknown–2 Aug 2025Alibaba (Qwen)Qwen-Imageimage generationopen weights permissive20BAug 2025Alibaba (Qwen)Wan 2.2video generationopen weights permissive–28 Jul 2025Alibaba (Qwen)Wan 2.2 A14Bvideo generationopen weights permissive27B28 Jul 2025Alibaba (Qwen)Wan 2.2 TI2V-5Bvideo generationopen weights permissive5B28 Jul 2025Alibaba (Qwen)Qwen3-Codercodeunknown–22 Jul 2025Alibaba (Qwen)Qwen3-Embedding-0.6Bembeddingopen weights permissive600MJun 2025Alibaba (Qwen)Qwen3-Embedding-4Bembeddingopen weights permissive4BJun 2025Alibaba (Qwen)Qwen3-Embedding-8Bembeddingopen weights permissive8BJun 2025Alibaba (Qwen)Qwen3-Reranker-0.6Brerankeropen weights permissive600MJun 2025Alibaba (Qwen)Qwen3-Reranker-4Brerankeropen weights permissive4BJun 2025Alibaba (Qwen)Qwen3-Reranker-8Brerankeropen weights permissive8BJun 2025Alibaba (Qwen)Qwen3LLMopen weights permissive–29 Apr 2025Alibaba (Qwen)Qwen3-0.6B-BaseLLMunknown600M28 Apr 2025Alibaba (Qwen)Qwen3-4B-BaseLLMunknown–28 Apr 2025Alibaba (Qwen)Qwen3-0.6BLLMopen weights permissive751.63M27 Apr 2025Alibaba (Qwen)Qwen3-1.7BLLMopen weights permissive1.7B27 Apr 2025Alibaba (Qwen)Qwen3-4BLLMopen weights permissive4B27 Apr 2025Alibaba (Qwen)Qwen3-8BLLMunknown8B27 Apr 2025Alibaba (Qwen)Qwen3-235B-A22BLLMopen weights permissive235.09B27 Apr 2025Alibaba (Qwen)Qwen3-32BLLMopen weights permissive32.76B27 Apr 2025Alibaba (Qwen)Qwen3-30B-A3BLLMopen weights permissive30.53B27 Apr 2025Alibaba (Qwen)Qwen2.5-Omni-7Bmultimodal LLMopen weights permissive10.73B22 Mar 2025Alibaba (Qwen)Wan-Fun (Wan2.1-Fun, Wan2.2-Fun)video generationopen weights permissive–17 Mar 2025Alibaba (Qwen)Wan2.1-VACE (1.3B, 14B)video generationopen weights permissive17.34B10 Mar 2025Alibaba (Qwen)QwQ-32Breasoningopen weights permissive32.5B6 Mar 2025Alibaba (Qwen)Wan 2.1 T2V 1.3Bvideo generationopen weights permissive1.42B25 Feb 2025Alibaba (Qwen)Qwen2.5-MaxLLMproprietary–28 Jan 2025Alibaba (Qwen)gte-reranker-modernbert-basererankeropen weights permissive149.61M20 Jan 2025Alibaba (Qwen)ZipEnhancermusic and audioopen weights permissive–9 Jan 2025Alibaba (Qwen)Qwen3-Coder-30B-A3B-Instructcodeopen weights permissive30.5B2025Alibaba (Qwen)Qwen3-Coder-480B-A35B-Instructcodeopen weights permissive480B2025Alibaba (Qwen)Wan 2.1video generationopen weights permissive–2025Alibaba (Qwen)QVQ-72B-Previewreasoningunknown73.41B24 Dec 2024Alibaba (Qwen)QwQ-32B-Previewreasoningopen weights permissive32.5B28 Nov 2024Alibaba (Qwen)Qwen2.5-Coder 14Bcodeopen weights permissive14B8 Nov 2024Alibaba (Qwen)Qwen2.5-Coder-32B-Instructcodeunknown32B6 Nov 2024Alibaba (Qwen)Qwen2.5-Coder 1.5Bcodeopen weights permissive1.5B18 Sep 2024Alibaba (Qwen)Qwen2.5-32B-InstructLLMunknown32B17 Sep 2024Alibaba (Qwen)Qwen2.5-0.5B-InstructLLMunknown–16 Sep 2024Alibaba (Qwen)Qwen2.5-72B-InstructLLMunknown72B16 Sep 2024Alibaba (Qwen)Qwen2.5-7B-InstructLLMunknown–16 Sep 2024Alibaba (Qwen)Qwen2.5-Coder 7Bcodeopen weights permissive7B16 Sep 2024Alibaba (Qwen)Qwen2.5-Math-1.5BLLMunknown–16 Sep 2024Alibaba (Qwen)Qwen2.5-Math-7BLLMunknown–16 Sep 2024Alibaba (Qwen)Qwen2.5 0.5BLLMunknown500M15 Sep 2024Alibaba (Qwen)Qwen2.5-14BLLMunknown–15 Sep 2024Alibaba (Qwen)Qwen2.5-32BLLMunknown32B15 Sep 2024Alibaba (Qwen)Qwen2-VL 2Bmultimodal LLMunknown2B5 Sep 2024Alibaba (Qwen)Qwen2-Audio-7B-Instructmultimodal LLMopen weights permissive8.4B31 Jul 2024Alibaba (Qwen)gte-multilingual-baseembeddingopen weights permissive305M20 Jul 2024Alibaba (Qwen)gte-multilingual-reranker-basererankeropen weights permissive305.96M20 Jul 2024Alibaba (Qwen)SenseVoiceSmallspeech recognitionopen weights restricted234M3 Jul 2024Alibaba (Qwen)gte-Qwen2-7B-instructembeddingopen weights permissive7.61B15 Jun 2024Alibaba (Qwen)Qwen2-7BLLMunknown–4 Jun 2024Alibaba (Qwen)UniAnimateavatarunknown–3 Jun 2024Alibaba (Qwen)emotion2vec+ largeaudio classificationopen weights restricted–15 May 2024Alibaba (Qwen)Qwen1.5-110B-ChatLLMopen weights restricted111.21B25 Apr 2024Alibaba (Qwen)gte-large-en-v1.5embeddingopen weights permissive434.14M20 Apr 2024Alibaba (Qwen)Qwen1.5LLMopen weights restricted–4 Feb 2024Alibaba (Qwen)Qwen1.5-72BLLMopen weights restricted72B4 Feb 2024Alibaba (Qwen)FunASR Conformer (English)speech recognitionopen weights restricted–1 Feb 2024Alibaba (Qwen)FSMN-VADaudio classificationopen weights permissive–1 Feb 2024Alibaba (Qwen)MossFormer2music and audioopen weights permissive–19 Dec 2023Alibaba (Qwen)I2VGen-XLvideo generationopen weights permissive1.42B7 Nov 2023Alibaba (Qwen)Qwen-14B-ChatLLMopen weights restricted14B24 Sep 2023Alibaba (Qwen)Qwen-VL-Chatmultimodal LLMunknown–20 Aug 2023Alibaba (Qwen)GTE (General Text Embeddings)embeddingunknown–27 Jul 2023Alibaba (Qwen)gte-largeembeddingopen weights permissive335.14M27 Jul 2023Alibaba (Qwen)gte-smallembeddingopen weights permissive33.36M27 Jul 2023Alibaba (Qwen)ERes2Netspeakeropen weights permissive–22 May 2023Alibaba (Qwen)ModelScope Text-to-Video (ModelScopeT2V)video generationopen weights restricted1.41B22 Mar 2023Alibaba (Qwen)FRCRNmusic and audioopen weights permissive–15 Jun 2022Alibaba (Qwen)Paraformer-zh (+ streaming)speech recognitionopen weights restricted220M2022Alibaba (Qwen)SIM (Search-based Interest Model)recommendationunknown–10 Jun 2020Alibaba (Qwen)DIN (Deep Interest Network)recommendationopen source–21 Jun 2017Alibaba (Qwen)CAM++speakeropen weights restricted––Alibaba (Qwen)Fun-ASR-MLT-Nanospeech recognitionunknown800M–Alibaba (Qwen)Lingjiang2.0time seriesunknown––Alibaba (Qwen)QwenLLMunknown––Alibaba (Qwen)Qwen TTS (Model Studio)speech synthesisunknown––Alibaba (Qwen)Qwen2LLMunknown––Alibaba (Qwen)Qwen2.5-Codercodeopen weights permissive––Alibaba (Qwen)Qwen3.5 Max PreviewLLMproprietary––Alibaba (Qwen)Qwen3.7 MaxLLMunknown––Alibaba (Qwen)Qwen3.7 Max PreviewLLMproprietary––Alibaba (Qwen)qwen3.7-plusLLMproprietary––Alibaba (Qwen)qwen3.8-flashLLMproprietary––Alibaba (Qwen)Qwen3.8-Max-PrimeLLMproprietary––Alibaba (Qwen)qwen3.8-omni-flashmultimodal LLMproprietary––Alibaba (Qwen)qwen3.8-omni-flash-realtimemultimodal LLMproprietary––Alibaba (Qwen)Qwen3-ASR-Flash-1208speech recognitionproprietary––Alibaba (Qwen)Wanvideo generationunknown––Alibaba (Qwen)Wan 2.5video generationproprietary––Alibaba (Qwen)Wan 2.6video generationproprietary––Alibaba (Qwen)Wan 2.7video generationproprietary––Alibaba (Qwen)Z-Image Baseimage generationunknown––

Scores

BenchmarkModelScoreWho ran itSetting
AIMEQwQ-32B-Preview50 %vendor-reported
AIME24Qwen2.5-32B-Instruct26.7 %vendor-reportedbaseline reported in the s1 paper
Arena Text leaderboardqwen3.7-plusrank onlyindependent
Arena Text leaderboardQwen3.8 Max1,483 Eloindependent
Arena Vision leaderboardQwen3.8 Max1,301 Eloindependent
Artificial Analysis Intelligence IndexQwen3.8 Max45 indexindependent0902
Artificial Analysis Intelligence IndexQwen3.8 Max45 indexindependentQwen3.8 Max (0902)
Artificial Analysis Intelligence IndexQwen3.8-2.4T-A95B40 indexindependent
Artificial Analysis Intelligence IndexQwen3.8-2.4T-A95B40 indexindependentQwen3.8 2.4T A95B
Artificial Analysis Text-to-Image ArenaQwen-Image-2.11,038 Eloindependentrank shown as 19-23; top open-weights model per AA FAQ
Artificial Analysis Text-to-Image ArenaQwen-Image-3.01,074 Eloindependent
Artificial Analysis Text-to-Image ArenaQwen-Image-3.0-Pro1,088 Eloindependentrank shown as 15-17
Artificial Analysis Text-to-Image ArenaZ-Image Turbo941 Eloindependentapproximate rank
Artificial Analysis Text-to-Speech ArenaQwen-Audio-3.1-TTS-Plus1,298 Eloindependent
Artificial Analysis Text-to-Video leaderboardHappyHorse 1.11,042 Eloindependent
Artificial Analysis Text-to-Video leaderboardWan 3.01,156 Eloindependent
C-EvalQwen1.5-72B84.1 %vendor-reported
CMMLUQwen1.5-72B83.5 %vendor-reported
FLEURSQwen3-ASR-0.6B7.92 WERvendor-reportedFLEURS Indonesian, Table A.2
FLEURSQwen3-ASR-1.7B5.16 WERvendor-reportedFLEURS Indonesian, Table A.2
FLEURSQwen3-ASR-Flash-12083.65 WERvendor-reportedFLEURS Indonesian, Table A.2
GIFT-EvalLingjiang2.00.457 CRPS (geo-mean, normalised to Seasonal Naive)independentzero-shot; rank is average CRPS rank
GIFT-EvalLingjiang2.00.666 MASE (geo-mean, normalised to Seasonal Naive)independentzero-shot
GPQAQwen3.5-397B-A17B88.4 %vendor-reported
GPQAQwQ-32B-Preview65.2 %vendor-reported
GPQA DiamondQwen3.5-9B81.7 %vendor-reportedNot independently verified
GPQA DiamondQwen3.8-2.4T-A95B92.6 %vendor-reported
GPQA DiamondQwen3.8-27B89.2 %vendor-reported
GPQA DiamondQwen3.8-Flash-Next91.7 %vendor-reported
GSM8KQwen1.5-72B79.5 %vendor-reported
Hugging Face Open ASR LeaderboardQwen3-ASR-0.6B6.42 WERmaintainer-runeight-set macro-average English WER
Hugging Face Open ASR LeaderboardQwen3-ASR-1.7B5.76 WERmaintainer-runeight-set macro-average English WER
HumanEvalQwen-14B-Chat43.9 %vendor-reported0-shot pass@1
HumanEvalQwen1.5-72B41.5 %vendor-reported
Humanity's Last Exam (HLE)Qwen3.8-2.4T-A95B43.6 %vendor-reported
LiveCodeBenchQwen3.8-27B90.3 %vendor-reported
LiveCodeBenchQwen3.8-27B90.3 %vendor-reportedLiveCodeBench v6
LiveCodeBenchQwen3.8-Flash-Next91.9 %vendor-reported
LiveCodeBenchQwQ-32B-Preview50 %vendor-reported
LMArena Text-to-Image ArenaQwen-Image-2.11,223 Eloindependentpreliminary
LMArena Text-to-Image ArenaQwen-Image-3.0-Pro1,255 Eloindependent
LMArena Text-to-Video ArenaHappyHorse 1.01,427 Eloindependent
LMArena Text-to-Video ArenaWan 2.71,337 Eloindependentwan2.7-t2v
LMArena Text-to-Video ArenaWan 3.01,476 Eloindependent
MATHQwen1.5-72B34.1 %vendor-reported
MATH500QwQ-32B-Preview90.6 %vendor-reported
MBPPQwen1.5-72B53.4 %vendor-reported
MMLUQwen-14B-Chat66.5 %vendor-reported5-shot
MMLUQwen1.5-72B77.5 %vendor-reported5-shot
MMLU-ProQwen3.5-397B-A17B87.8 %vendor-reported
MMLU-ProQwen3.5-4B79.1 %vendor-reportedNot independently verified
MMLU-ProQwen3.5-9B82.5 %vendor-reportedNot independently verified
MMLU-ProQwen3.6-27B86.2 %vendor-reported
MMLU-ProQwen3.7 Max89.6 %vendor-reported
MMTEB-R (reranking)Qwen3-Reranker-8B72.94 scorevendor-reported
MT-BenchQwen1.5-110B-Chat8.88 scorevendor-reportedaverage score
MTEB multilingualQwen3-Embedding-8B70.58 scorevendor-reported
MTEB-R (reranking)Qwen3-Reranker-8B69.02 scorevendor-reported
OSWorld-VerifiedQwen3.8 Max86.1 %vendor-reportedThe genealogy notes label the holder open weights; not supported by its sources
Qwen-Image-BenchQwen-Image-2.160.28 scorevendor-reportedQwen chart only; methodology unreleased
Qwen-Image-BenchQwen-Image-3.0-Pro62.36 scorevendor-reportedQwen chart only; methodology unreleased
SEA-HELMQwen3-8B60.33 scoremaintainer-runCheckpoint Qwen3-8B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3-8B69.36 scoremaintainer-runCheckpoint Qwen3-8B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3-VL-8B-Instruct56.6 scoremaintainer-runCheckpoint Qwen3-VL-8B-Instruct as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3-VL-8B-Instruct67.46 scoremaintainer-runCheckpoint Qwen3-VL-8B-Instruct as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.5-122B-A10B74.66 scoremaintainer-run
SEA-HELMQwen3.5-122B-A10B77.93 indexmaintainer-runIndonesian
SEA-HELMQwen3.5-122B-A10B77.93 scoremaintainer-run
SEA-HELMQwen3.5-35B-A3B73.55 scoremaintainer-runCheckpoint Qwen3.5-35B-A3B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.5-35B-A3B77.55 scoremaintainer-runCheckpoint Qwen3.5-35B-A3B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.5-4B63.82 scoremaintainer-runCheckpoint Qwen3.5-4B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.5-4B69.66 scoremaintainer-runCheckpoint Qwen3.5-4B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.5-9B69.13 scoremaintainer-runCheckpoint Qwen3.5-9B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.5-9B74.23 scoremaintainer-runCheckpoint Qwen3.5-9B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.6-27B74.64 scoremaintainer-run
SEA-HELMQwen3.6-27B77.96 indexmaintainer-runIndonesian
SEA-HELMQwen3.6-27B77.96 scoremaintainer-run
SEA-HELMQwen3.6-35B-A3B73.86 scoremaintainer-runCheckpoint Qwen3.6-35B-A3B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.6-35B-A3B77.29 scoremaintainer-runCheckpoint Qwen3.6-35B-A3B as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.8-Flash-Next73.04 scoremaintainer-runCheckpoint Qwen3.8-Flash-Next as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SEA-HELMQwen3.8-Flash-Next79.04 scoremaintainer-runCheckpoint Qwen3.8-Flash-Next as listed; release view (data last updated 18 Sep 2026); score is the average of 2000 bootstraps
SWE-bench ProQwen3-Coder-480B-A35B-Instruct38.7 %vendor-reported
SWE-bench ProQwen3-Coder-Next44.3 %vendor-reportedscaffold not stated
SWE-bench ProQwen3.8-2.4T-A95B67.7 %vendor-reported
SWE-bench ProQwen3.8-27B61.7 %vendor-reported
SWE-bench ProQwen3.8-27B61.7 %vendor-reported
SWE-bench ProQwen3.8-Flash-Next62.5 %vendor-reported
SWE-bench VerifiedQwen 3.6 Plus78.8 %vendor-reported
SWE-bench VerifiedQwen3-Coder-Next70.6 %vendor-reportedscaffold not stated
SWE-bench VerifiedQwen3.5-397B-A17B76.4 %vendor-reported
SWE-bench VerifiedQwen3.6-27B77.2 %vendor-reported
SWE-bench VerifiedQwen3.6-35B-A3B73.4 %vendor-reported
Terminal-BenchQwen3-Coder-480B-A35B-Instruct23.9 %vendor-reported
Terminal-BenchQwen3-Coder-Next36.2 %vendor-reportedscaffold not stated
Terminal-BenchQwen3.6-35B-A3B51.5 %vendor-reported
Terminal-BenchQwen3.8-2.4T-A95B86.6 %vendor-reported
Terminal-BenchQwen3.8-27B73 %vendor-reportedTerminus
VBenchWan 2.186.22 %maintainer-runTotal score; submission-based board
VBenchWan 2.186.67 %maintainer-runQuality score; submission-based board
Where this comes from
Every recorded score of this organisation's models, with who ran it. Scores from different settings are not directly comparable.

Sources

  1. Qwen3.8 Max (0902) - Intelligence, Performance & Price Analysis | Artificial Analysis artificialanalysis.ai https://artificialanalysis.ai/models/qwen3-8-max
  2. Recommended models - Alibaba Cloud alibabacloud.com https://www.alibabacloud.com/help/en/model-studio/models
  3. Qwen (Qwen) huggingface.co https://huggingface.co/Qwen
  4. Gladia - Best open-source speech-to-text models in 2026 gladia.io https://www.gladia.io/blog/best-open-source-speech-to-text-models
  5. GitHub - modelscope/FunASR: Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diariza github.com https://github.com/modelscope/FunASR
  6. Alibaba Group - Wikipedia en.wikipedia.org https://en.wikipedia.org/wiki/Alibaba_Group
  7. Qwen 3.5 Small Models: 9B Parameters That Beat 120B rits.shanghai.nyu.edu https://rits.shanghai.nyu.edu/ai/qwen-3-5-small-models-9b-parameters-that-beat-120b
  8. 9个月内用户增长千万 魔搭社区已服务2500万开发者 stdaily.com https://www.stdaily.com/web/gdxw/2026-03/22/content_490099.html
  9. How we built a custom vision LLM to improve document processing at Grab engineering.grab.com https://engineering.grab.com/custom-vision-llm-at-grab
  10. aisingapore/Qwen-SEA-LION-v4.5-27B-IT , Hugging Face huggingface.co https://huggingface.co/aisingapore/Qwen-SEA-LION-v4.5-27B-IT
  11. Qwen3 Technical Report arxiv.org https://arxiv.org/abs/2505.09388

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Alibaba (Qwen)", https://aurion.semai.id/organisations/alibaba/, records of October 2026.