Qwen · Text
Qwen models: Versions, specs, GPU requirements & hosting
Qwen is a broad ecosystem whose general, coder, vision, omni and embedding branches are tracked independently instead of forced into one timeline.
Overview and history
Qwen is an official Qwen model family represented here through independently verified release, checkpoint, license, runtime and hardware evidence. Qwen is a broad ecosystem whose general, coder, vision, omni and embedding branches are tracked independently instead of forced into one timeline. Current public releases are grouped by the General, Coder, Vision, Omni, Embedding, Fast Multimodal tracks so sibling sizes and specialist branches are not presented as false chronological generations.
FusionPointAI currently tracks 99 certified public Qwen releases. The catalogue preserves release dates and evaluates Current, Recent, Previous and Legacy status within each model track rather than by database insertion order.
Broad modality coverage
Dense and MoE variants
Strong deployment ecosystem
Architecture and model range
The verified public range spans 8B to 2400B total parameters and context windows up to 262,144 tokens. Each release states whether it is dense or mixture-of-experts and records active parameters separately when the official source publishes them.
Public releases99
Model tracks5
Current generationQwen3.8-Flash-Next
Last verified2026-09-04T00:00:00.000000Z
Generations, tracks and variants
General
contextpilot-8b
Recent generationqwen3
Recent generation 14 sibling variants千问3-235B-A22B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
千问3-8B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
千问3-30B-A3B-Instruct-2507
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3-8B-AWQ
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Unknown / Not Recommendable
Qwen3-235B-A22B-GPTQ-Int4
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Unknown / Not Recommendable
Qwen3-8B-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Unknown / Not Recommendable
Qwen3-235B-A22B-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Unknown / Not Recommendable
Qwen3-30B-A3B-Instruct-2507-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-235B-A22B-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3Guard-Stream-8B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Fine-tune
Qwen3Guard-Gen-8B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Fine-tune
Qwen3 8B
8.0B parameters · 40,960 context
Dense
Qwen3ForCausalLM
Qwen3 235B-A22B
235.0B parameters · 40,960 context
MoE
Qwen3MoeForCausalLM
Qwen3 30B-A3B Instruct 2507
30.5B parameters · 262,144 context
MoE
Qwen3MoeForCausalLM
qwen3-8b
Recent generation 2 sibling variantsqwen3-235b-a22b
Recent generationqwen3-8b-dms-8x
Recent generationnemotron-8
Recent generationintent-aware-lfqa-qwen3-8b-multiview
Recent generationqwen3-235b-a22b-eagle3
Recent generationintent-aware-lfqa-qwen3-8b-intent-implicit
Recent generationintent-aware-lfqa-qwen3-8b-baseline
Recent generationwebworld-8b
Recent generationwebworld-32b
Recent generationwebworld-14b
Recent generationmolmopoint-8b
Recent generationintent-aware-lfqa-qwen3-8b-intent-explicit
Recent generationpenguin-vl-8b
Recent generationmolmoweb-pretrained-8b
Recent generationmolmo2-8b
Recent generationmolmoweb-4b
Recent generationwedlm-8b-base
Recent generationmolmoweb-8b
Recent generationnemotron-orchestrator-8b
Recent generationmolmoweb-4b-native
Recent generationcudallm-8b
Recent generationmolmoweb-8b-native
Recent generationqwen3-8-2-4t-a95b
Recent generationmolmoweb-pretrained-4b
Recent generationtmax-sft-8b
Recent generationqwen3-8
Current generation 6 sibling variantsQwen3.8-2.4T-A95B-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Unknown / Not Recommendable
Qwen3.8-27B-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3.8-2.4T-A95B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Unknown / Not Recommendable
千问3.8-27B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3.8 27B
27.0B parameters · 262,144 context
Dense
Qwen3_5ForConditionalGeneration
Qwen3.8 2.4T-A95B
2,400.0B parameters · 262,144 context
Flagship · MoE
Qwen3_5MoeForCausalLM
Code
qwen3
Recent generation 3 sibling variants千问3-Coder-Next-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3-Coder-Next
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
千问3-Coder-Next-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3 Coder Next
Current generationVision
sage-mm-qwen3-vl-8b-sft
Recent generationvisual-erm
Recent generationqwen3-vl-235b-a22b-instruct
Recent generationegm-8b-sft
Recent generationegm-8b
Recent generationqwen3
Recent generation 26 sibling variantsQwen3-VL-235B-A22B-Instruct-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-Embedding-8B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Fine-tune
Qwen3-VL-235B-A22B-Thinking-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3-VL-8B-Thinking
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
千问3-VL-32B-Thinking
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
千问3-VL-8B-Instruct
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
千问3-VL-30B-A3B-Instruct
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3-VL-30B-A3B-Thinking-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-32B-Thinking-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-8B-Thinking-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-30B-A3B-Instruct-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-8B-Instruct-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-235B-A22B-Thinking-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-235B-A22B-Instruct-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-32B-Instruct-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-8B-Thinking-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-32B-Thinking-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Qwen3-VL-8B-Instruct-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3-VL-32B-Instruct
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3-VL-32B-Instruct-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3-VL-235B-A22B-Instruct
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3-VL-30B-A3B-Instruct-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3-VL-30B-A3B-Thinking
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3-VL-30B-A3B-Thinking-GGUF
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
千问3-VL-235B-A22B-Thinking
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Qwen3-VL-Reranker-8B
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Fine-tune
cosmos-reason2-32b
Recent generationcosmos-reason2-8b
Recent generationunirg-cxr
Recent generationQwen3-VL
Current generation 8 sibling variantsQwen3-VL 235B-A22B Thinking
235.0B parameters · 262,144 context
Vision · Reasoning · MoE
Qwen3VLMoeForConditionalGeneration
Qwen3-VL 30B-A3B Thinking
30.5B parameters · 262,144 context
Vision · Reasoning · MoE
Qwen3VLMoeForConditionalGeneration
Qwen3-VL 235B-A22B Instruct
235.0B parameters · 262,144 context
Vision · MoE
Qwen3VLMoeForConditionalGeneration
Qwen3-VL 30B-A3B Instruct
30.5B parameters · 262,144 context
Vision · MoE
Qwen3VLMoeForConditionalGeneration
Qwen3-VL 8B Thinking
8.0B parameters · 262,144 context
Vision · Reasoning · Dense
Qwen3VLForConditionalGeneration
Qwen3-VL 8B Instruct
8.0B parameters · 262,144 context
Vision · Dense
Qwen3VLForConditionalGeneration
Qwen3-VL 32B Thinking
32.0B parameters · 262,144 context
Vision · Reasoning · Dense
Qwen3VLForConditionalGeneration
Qwen3-VL 32B Instruct
32.0B parameters · 262,144 context
Vision · Dense
Qwen3VLForConditionalGeneration
Omni
qwen3
Recent generationQwen3-Omni
Current generationFast multimodal
qwen3-8
Recent generation 2 sibling variantsQwen3.8-Flash-Next-FP8
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Quantization
Unknown / Not Recommendable
千问3.8-Flash-Next
The manufacturer has not published this information. parameters · N/A / Not directly comparable context
Base
Unknown / Not Recommendable
Qwen3.8 Flash Next
Current generationDeployment ecosystem
Choose a specific Qwen release and certified checkpoint before sizing hardware. FusionPointAI rejects unknown runtime configurations, calculates resident memory from total model weights, then distinguishes globally compatible GPUs from currently rentable inventory.
Validate the exact checkpoint license, runtime version, precision kernel, context length, concurrency and multi-GPU topology. Active MoE parameters affect compute, but resident model memory continues to use total stored weights.
Common use cases
Multilingual assistants
Vision-language services
Coding agents
License overview
Licenses are certified per release, not inferred from the provider. Review the release page before commercial deployment because terms can differ between generations and variants.
Commercial-use status is a source-backed catalogue classification, not legal advice. Follow the linked official license and attribution requirements.
Release timeline
qwen/qwen/3-coder-next-gguf @ 9ba46fd174edfac1f66d16331bd6094978d69d25440968942c6626c7283a74b2
qwen/qwen/3-coder-next-fp8 @ 56efdea522c86828594edd12d4f921ce369426834df1c7d19f4a6cf9c6b4acb8
qwen/qwen/qwen3-vl-8b-thinking-fp8 @ c8cdffd1e2f526a5a79981d9892b09f405d07ee8ca936ab8fa079f8021ffbd0c
qwen/qwen/qwen3-235b-a22b-gguf @ e2865c756c5484eacc7a060254b007a9bbbe99d85bca1bb8b1138af230f599f5
qwen/qwen/qwen3-vl-235b-a22b-thinking-fp8 @ ad94fb393f290dd06a39ee41bc7b9c0c75d23798ea0f2e5eb7bfcf342bfbb1a9
qwen/qwen/qwen3-30b-a3b-instruct-2507-fp8 @ cb8d9632954dd14a296154174d3d1b5bd7b32ea36bd6ef3f293bb491f3e33553
qwen/qwen/qwen3-vl-235b-a22b-instruct-fp8 @ 14d6cfdb5445699e1c5fd1484430cd19e2696f5d7eabddec79066153573f7e94
qwen/qwen/qwen3-vl-30b-a3b-instruct-fp8 @ 59cdd8c66d9533417ab19426739d79ceaa87646647a0fcceb1d4ad717ae17ed8
qwen/qwen/qwen3-vl-30b-a3b-instruct-gguf @ 42421fb0a59b939ee4ac831958ad976a8919d4503785ff094f9aa3deaf639a4d
qwen/qwen/qwen3-vl-32b-instruct-gguf @ 9a96dcea5829779d797dba61c1ef462dc09cbd94e3de3a09f6cfdf453c7338df
qwen/qwen/qwen3-vl-32b-instruct-fp8 @ 78797c3a6d119c339158a97f6cbb6606435b7fbb649434b7281e36187839bb42
qwen/qwen/qwen3-8b-fp8 @ 08b7fc851a62e6e55ce2f63dd4055fb226263302e469da51280bf35b701fd9b8
qwen/qwen/qwen3-vl-8b-instruct-fp8 @ 085fcdcfef7ff37cda66d4ec2f15adbd5e2413946590a2c12ef1f618ba7bb888
qwen/qwen/qwen3-8b-awq @ 3a23ef06da2f0dab50a69e7ab287233765b94744b954d8fd0e55f70e5964bba9
qwen/qwen/qwen3-vl-8b-instruct-gguf @ 662cae7e77338b4054df4742d62654490e59beb520217c0021fb92edab876469
qwen/qwen/qwen3-8b-gguf @ 390a4e43c04835db7680fe017e7e3298eaa032b8efcb1232b67a425eb99f4f3b
qwen/qwen/qwen38-flash-next-fp8 @ 9115325ef004e28697ebd2d1bde88775e11289a6735d54e1579bf361b9c3f447
qwen/qwen/qwen38-27b-fp8 @ bb05758dbab17b847b1b6055eb410a11094093347fe4c970ef3f4ea55275c969
qwen/qwen/qwen38-24t-a95b-fp8 @ 88ac67975a942d347ecf27f676232d1f0e37b7cd7ec7c80f592deffcbd50279c
qwen/qwen/qwen3-235b-a22b-gptq-int4 @ 9f5c3c120042637c410db1274e20c81252dc61525e242584dd18d54fd527460a
qwen/qwen/qwen3-235b-a22b-fp8 @ 2ce032b662018a9409ca2168817c9b13004dbc4a768522289b0ea6c7f9f4c8a3
qwen/qwen/qwen3-235b-a22b-nvfp4 @ fd8025127c6f90cad946096d84aa218a67ecb5448f5620daaa9c192320bd8c87
qwen/qwen/3-235b-a22b @ 05e29eac5554bb5275ab8cc8a47c918b969f444dd9768505149f6b80455b8718
qwen/qwen/cudallm-8b @ 200bdd1b97d21e24cac3caae7b4e17d1c87ea031792dd141739e5196cbdf1d57
qwen/qwen/qwen3-8b-nvfp4 @ 7531b4ee9ac298e8544413ff67158ed9b5efff76e2ccb3c07f82f439870d90a0
qwen/qwen/3-30b-a3b-instruct-2507 @ 3457e42b13ff827ce034ace2914fb88b93b4b99c1e180313062410716898224f
qwen/qwen/3-omni-30b-a3b-instruct @ df1c6e7c2b3e3fdd476f7a12c7a3fd737c37c0b2eebc7138e881ec4ba283440c
qwen/qwen/3-vl-8b-instruct @ 911389e8b11b10b86d4bbd5363b4c97cd75df1add29552e868472afde19b2214
qwen/qwen/3-vl-32b-instruct @ b3f0f78a6cae5fd719088a551bbeb3f11f95b751f9043ca173b136d77e098c00
qwen/qwen/3-vl-32b-thinking @ 69739e9134e417a60d1923257cee5b231761ccf5f7146db0de1f8c3f035969b8
qwen/qwen/qwen3-vl-235b-a22b-instruct-gguf @ b79f38a050df467e4d04f3f0d7fb93965d12841896226ac44be84fce20b5a575
qwen/qwen/qwen3-vl-30b-a3b-thinking-gguf @ d92b9afa8d72358f026f8d19f8563e5abbfd0360ae55c665de0b9c799752a970
qwen/qwen/qwen3-vl-235b-a22b-thinking-gguf @ 8039f52c056caa3f0ece7dbc7408d8074e79498fc1a7cd79996158d2a42ef852
qwen/qwen/qwen3-vl-32b-thinking-gguf @ ef2ef2c86f13bbd506b334f2195abc9196d7f5f19a7d7934a1c6a13963b75d5b
qwen/qwen/qwen3-vl-8b-thinking-gguf @ 79a34477e21735a6bd5b552f81cce294ae4a3473cf46817f9a4fff174e517a86
qwen/qwen/qwen3guard-stream-8b @ 9ef13e320b9903036c235d3140190c8e7abf027679bf2ec976106e329ce6b4ca
qwen/qwen/qwen3guard-gen-8b @ 5f8b6c2d4223456c3cef3328f07dd49d67caa5e0c9e19af23df25a104116da08
qwen/qwen/3-vl-30b-a3b-thinking @ fc1e9559a1c7e3651d75dc087ee8ab58a259c5d027e9a09cf0a49d47a1f64e74
qwen/qwen/3-vl-30b-a3b-instruct @ 6d8f380582847d8f281580a9cb004d6b623377b9e63ae3fecc1629de3e7f8fa6
qwen/qwen/qwen3-vl-30b-a3b-thinking-fp8 @ fb3086e1611a6cc53f536dc4caaba53f89a214ae60aad939a5332285b1c4331f
qwen/qwen/3-vl-8b-thinking @ b60ee8668038ea3298067013ba869ba62f0a93b7b8c650cceaaed6c8e4b684ce
qwen/qwen/3-vl-235b-a22b-instruct @ 0a59cd4a764d7e8a6d3409cd08a88c18af531760f0b5dd60d8fb781c8977486f
qwen/qwen/3-vl-235b-a22b-thinking @ 17273f7a651e896d4e01977d8ceb70854371bcd2b9111019556c284de89e979c
qwen/qwen/nemotron-orchestrator-8b @ efc22ef7083eb63bb7a7097a4bc2c3d9fff8ccade392a38858d432f1dda7af42
qwen/qwen/sage-mm-qwen3-vl-8b-sft @ 1fca68df88a473d502af17744dbbd9c3768e11b38a4e686a1d421171070a49c6
qwen/qwen/qwen3-nemotron-8b-brrm @ fc7b943bbfb803601ebb5212009ee70b08622656e025d344c6e1e7a32ab25421
qwen/qwen/wedlm-8b-base @ 2721c402a53e73d3618d86c0e35363d8906d33d84f104287aed38ae60431e835
qwen/qwen/qwen3-8b-dms-8x @ 61d84f454f2e25f2fc021fff3111a98678206608d4a727d5d2916dba13d7d23f
qwen/qwen/molmo2-8b @ 7fd96f388787d73cff9a004f22d418e73cd30ccb358215c4254cfc181cd65c78
qwen/qwen/qwen3-235b-a22b-eagle3 @ 90e6ab46de62baec3ddaa05023a4f7cd3b501344bf8cdc17c83690399260dbd0
qwen/qwen/3-coder-next @ 8e3b0e58d5c441d8a7e1530390c3012004fd8a4662dce3f145a5cd1ae463a562
qwen/qwen/qwen3-vl-235b-a22b-instruct-nvfp4 @ d9035575439f2f50147bd3d4e2f459b40941d02bcb3a8b7991adc2dc59962935
qwen/qwen/penguin-vl-8b @ b3748eea83aa5bdbbcfa66c10e713763c1382b4456ee283f4b5dc0938f0f4229
qwen/qwen/molmopoint-8b @ c243e1574f54564ac2519f2524efc25ea4b7425dd48792ec814c0d22e84c31fe
qwen/qwen/visual-erm @ 9e0631650ebc5580e26b6c5a2417becbd7c81a36015c85b668bfdd112077680b
qwen/qwen/egm-8b @ f4d9ec63d886d013e3f65a22ca275b65e92d3e5056f82b40fe96c3f05a90810d
qwen/qwen/molmoweb-pretrained-4b @ 6e41879eff53bd3ed3be75c8bb21119763391c85aceafa61dc7f349f49153f0f
qwen/qwen/egm-8b-sft @ d7c2d5370f3802ad67606e0bdecb429289c0346f639713888e1fcd20409d3c8c
qwen/qwen/molmoweb-4b @ 768611171bbbbcdedfff5bc1710a7d2ec5e1cea30fad0db38a27a6057d2836ea
qwen/qwen/molmoweb-8b @ ac05472e1ceca192ca38c75e05047e6efb301129f5e685759bab213cbb54ef01
qwen/qwen/molmoweb-4b-native @ e37718c2f5b40ecb01261a270ec81e30bc88b9ecebb4600a9a142c0e75ceb0be
qwen/qwen/molmoweb-8b-native @ f832353b2e80b898c2978e1fd19729e2034d6410a3211775a3315723fbf26e2a
qwen/qwen/molmoweb-pretrained-8b @ 097362abde107fa79776f638739ef7dff09b7a2b95f41111b6393ac2a204a10b
qwen/qwen/qwen3-vl-reranker-8b @ 0e26196df977407f276bae4f03e5b4d106bc61fc5061dce8027e8560970f3d71
qwen/qwen/qwen3-vl-embedding-8b @ 5b4bbaf0d86d02c2c466475e7d09c7b8f607fc438282df2a0c473097cdc61382
qwen/qwen/intent-aware-lfqa-qwen3-8b-intent-implicit @ 1088309ef23958453bd492c0880d8f73c8c6eef6e55866d04d2d0fddeb5eb0f3
qwen/qwen/intent-aware-lfqa-qwen3-8b-baseline @ 182475a35c9f77932ea556f4827fb290dfc8095bd0caf8c36e916af966dbfcac
qwen/qwen/intent-aware-lfqa-qwen3-8b-multiview @ 518a110a1e7cdb61c68f0a5b6c3f5f37758859d649fc99046affa33f5da4b234
qwen/qwen/intent-aware-lfqa-qwen3-8b-intent-explicit @ fccfb3be57d201282048087ad139f0297b54cda7a668938b3fe68f73e583bcd5
qwen/qwen/cosmos-reason2-32b @ a91cdfc0642b0fb96e9b266dabff8fb240c05f6364c98289788d1ea05c27f6b0
qwen/qwen/cosmos-reason2-8b @ cec522e649323044a0bc09e835ecb4f2e5df33d3bbb741f30cf68a74759baad3
qwen/qwen/unirg-cxr @ 52f86194b82ba610729641d434b9c1322cc912e3cfa512ec8b64804214b67a8f
qwen/qwen/webworld-14b @ f3bcac4d7c42a5c165d7032089f3470c15311da2ebc0c3e81607ad9f63e0eb63
qwen/qwen/webworld-32b @ 004a7441102771dc72d30ce38405ff4a5c1def931bdfcda6bd5562a0a602e048
qwen/qwen/webworld-8b @ d3b2bdb7d67f3880c9312d804183907d447ba2c39cb804e3c074d5e0c155632f
qwen/qwen/tmax-sft-8b @ 0c3b8fd3b1f493f5af02d4ec9ed77ca62b75008a12b501cf78f446b8e1583f52
qwen/qwen/38-24t-a95b @ 5f21298489ad09956a4bf62f01eacbddb379e4f1e9839ecdaeace8b064ccbdea
qwen/qwen/38-27b @ f1e2be0af2dcbb9fd8aa217277440c9bfc2c87fc89eaaccd4e38a9e914080201
qwen/qwen/38-flash-next @ 1acb09cfd67423dfa0dece2803b39f83032d84db051b3a5e969f7858a466b751
qwen/qwen/contextpilot-8b @ 8b73d630ab2b7576fa34baf5b089ea6425cda810cc1267c34563816f89da5473
qwen/qwen/qwen38-24t-a95b-nvfp4 @ c70bacf064d93a471ad2eba2cb8a7971fe42bb23e29f57259cbbae543f85b7a0
qwen/qwen/qwen3-30b-a3b-instruct @ 329411abe27fa043aad38ecf002e0a4a193baf50ce0e3f9af9a03ca23df423a2
qwen/qwen/qwen3-235b-a22b @ fd813fb51be324f0967d4ce6f7b2d172908d74455e1aa47ff16d263f5cfdff85
qwen/qwen/qwen3-8b @ 872a82fc823d6d3fc1ec27684ffa4391e0c2b8b6662cae48c7d453b573800113
qwen/qwen/qwen3-vl-8b @ a1cf7a83c85ddba10a0cf162de2cdefe8ee46c45c346c1b0833457a9d22c5ba8
qwen/qwen/qwen3-8-27b @ 3e7c75625b03cf6f2d7256b8dd37aae1edd2bdfed8b6489d80f23c9a09db6165
qwen/qwen/qwen3-8-2-4t-a95b @ 74e316d144b8d12c4c435d77676e436595e7235267d84610e0c3d17b49399151
qwen/qwen/qwen3-coder-next @ a7b14ccb2fe1311a2b86f746a4c9703a16ca8027f692c1ba7a61ebb94fce202a
qwen/qwen/qwen3-vl-235b-a22b-instruct @ f48309b67419fb01ac97e977bdcb0027ca5f857e7be798994dbad0e42e01d7fd
qwen/qwen/qwen3-vl-235b-a22b-thinking @ efacc98c801e9cb648a18ec5fea1b6d65af562a05f6589735cb316019a91688d
qwen/qwen/qwen3-vl-32b-instruct @ 93d6b6b8fdd4c11176f12b21c22f5123de4357afef6dcf3de201541f930d2c53
qwen/qwen/3-8b @ 4b285c1eb23379708ee18e5dc008b471c298c71fd13ad422f9945e26bc01c5e3
qwen/qwen/qwen3-vl-32b-thinking @ fd1c68f471a54c8703a337f2cc47090ee2ab85220d447bb57cdcebf793794d4e
qwen/qwen/qwen3-vl-30b-a3b-instruct @ 22a551d409dbe1c42a25204a48587c2336be075472f27d6879f32386d65dec31
qwen/qwen/qwen3-vl-30b-a3b-thinking @ 3c291564381627986e8cacce990e0f5326072954c047db2f29f31a96e6029e84
qwen/qwen/qwen3-vl-8b-thinking @ efcf28be5e5b19b8e5ffeab629393ec8d0873df819088ff4c4a31dfc17abb279
qwen/qwen/qwen3-omni-30b-a3b-instruct @ 760edcedc32f3a85f57a3c45e9b43771ae1f57dc4210ad48ae710ac08303cf7b
qwen/qwen/qwen3-8-flash-next @ 24b6d65850ab647254a59896ca97b9294561271ff81fa011700f4fe01c4adbe2
Qwen/Qwen3.8-Flash-Next-FP8 @ 2026-08-31T08:24:29Z
Frequently asked questions
Which Qwen release is current?
Current releases are Qwen3-VL 8B Instruct, Qwen3.8 27B, Qwen3.8 2.4T-A95B, Qwen3 Coder Next, Qwen3-VL 235B-A22B Instruct, Qwen3-VL 235B-A22B Thinking, Qwen3-VL 32B Instruct, Qwen3-VL 32B Thinking, Qwen3-VL 30B-A3B Instruct, Qwen3-VL 30B-A3B Thinking, Qwen3-VL 8B Thinking, Qwen3-Omni 30B-A3B Instruct, Qwen3.8-Flash-Next, 千问3.8-27B, 千问3.8-2.4T-A95B, evaluated within their respective tracks.
Can Qwen be self-hosted?
Only releases with a verified official checkpoint and supported runtime configuration are presented as deployable.
Which GPU should run Qwen?
GPU selection depends on the exact variant, checkpoint precision, context length and concurrency; use the release hardware recommendations rather than the family name alone.
Official references
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- modelscope.cn
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- huggingface.co
- modelscope.cn
- modelscope.cn
- huggingface.co
Last verified: 2026-09-04