Up to date
Qwen tongyi-qianwen-license-agreement out 3 years ago
7.72B parameters, text in, text out, 32K tokens of context.
Verified 2026-10-25 Rank #312 in LLMs · score 83 method v1 Sources 3 Downloads 30d 46.3k · #391 of 619 tracked
Picked by rule from 35 tracked pieces: the lab's own post, the most shared, a field report, the latest.
Strata runs the 125-billion-parameter Qwen3.8-Flash-Next model on consumer hardware under an MIT license with no cloud dependencies.
The 2 picks above, plus these 23. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
No API provider serves it, so its downloads are people running it themselves: 8 quantized builds and 100 Spaces.
Top Spaces running it: LanguageBind/MoE-LLaVA ♥ 154 · EmbeddedLLM/chat-template-generation ♥ 12 · Justinrune/LLaMA-Factory ♥ 12 · lightmate/llm-chatbot ♥ 9
Weights only, an estimate: leave room for the KV cache. Method
About 4.6 GB at INT4, 9.3 GB at FP8, 19 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2023-08-03), amber triangles are 10 dated articles (each links to it), grey ticks are repo commits.
22 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in LLMs
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | QWenLMHeadModel |
| Model type | qwen |
| Params by dtype | BF16: 7,721,324,544 |
| Size band | M |
| MoE | no |
| Context length | 32,768 (config.json max_position_embeddings) |
| Languages | zh, en |
| Papers | arXiv:2309.16609, arXiv:2305.08322, arXiv:2009.03300, arXiv:2305.05280, arXiv:2210.03629 |
| License | tongyi-qianwen-license-agreement (other) |
| Hub snapshot | 2026-10-25 |
vLLM: not in registry · QWenLMHeadModel · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · QWenLMHeadModel · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for Qwen-7B-Chat: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 19 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 9.3 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 4.6 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 8finetune: 1adapter: 13
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| TheBloke/Qwen-7B-Chat-AWQ | quantized | AWQ | 7.72B | M | — | TheBloke | 9 | 28.8k |
| TheBloke/Qwen-7B-Chat-GPTQ | quantized | GPTQ | 7.72B | M | — | TheBloke | 3 | 9.3k |
| mradermacher/Qwen-7B-Chat-i1-GGUF | quantized | GGUF | — | — | other | mradermacher | 1 | 6.5k |
| Xorbits/Qwen-7B-Chat-GGUF | quantized | GGUF | — | — | — | Xorbits | 8 | 4.7k |
| mradermacher/Qwen-7B-Chat-GGUF | quantized | GGUF | — | — | other | mradermacher | 1 | 3.8k |
| stvlynn/Qwen-7B-Chat-Cantonese | finetune | — | 7.72B | M | agpl-3.0 | stvlynn | 23 | 2.2k |
| h31337/Qwen-7B-Chat-GGUF | quantized | GGUF | — | — | — | h31337 | 0 | 925 |
| Jungwonchang/Ko-QWEN-7B-Chat-LoRA | adapter | — | — | — | — | Jungwonchang | 7 | 432 |
| mlc-ai/Qwen-7B-Chat-q4f16_1-MLC | quantized | — | — | — | — | mlc-ai | 2 | 275 |
| xjw1001002/Qwen_match_price_V2 | adapter | — | — | — | other | xjw1001002 | 0 | 151 |
| yooshijay/qwen-7B_psychat | adapter | — | — | — | — | yooshijay | 0 | 99 |
| ggkk2012/qwen7b-finance | adapter | — | — | — | other | ggkk2012 | 1 | 49 |
| lyy14011305/firefly-qwen-7b-sft-qlora | adapter | — | — | — | — | lyy14011305 | 0 | 98 |
| mlc-ai/Qwen-7B-Chat-q4f32_1-MLC | quantized | — | — | — | — | mlc-ai | 0 | 92 |
| hollywang/wangxf.holly-Qwen-7B-LoRA_demo | adapter | — | — | — | apache-2.0 | hollywang | 1 | 41 |
| qiyongli22/Test2 | adapter | — | — | — | — | qiyongli22 | 0 | 80 |
| xjw1001002/price_tag_V3_qwen7b | adapter | — | — | — | other | xjw1001002 | 0 | 66 |
| qiyongli22/Qwen_test | adapter | — | — | — | — | qiyongli22 | 0 | 64 |
| xjw1001002/Qwen7B_brand_model | adapter | — | — | — | other | xjw1001002 | 0 | 45 |
| kyriemao/chatretriever | adapter | — | — | — | — | kyriemao | 0 | 36 |
| Zetsu00/qwen-7b-lora-simplifier-v2 | adapter | — | — | — | — | Zetsu00 | 0 | 33 |
| Zetsu00/Qwen-7B-QLoRA-simplifier | adapter | — | — | — | — | Zetsu00 | 0 | 17 |
| Date | Commit | SHA |
|---|---|---|
| 2024-03-19 | Update README.md (#53) | 93a65d3 |
| 2024-01-04 | add license | 8d24619 |
| 2023-12-13 | update wechat | 218aa32 |
| 2023-12-13 | update wechat | 555e901 |
| 2023-12-07 | update modeling_qwen.py | be72f02 |
| 2023-12-06 | update modeling_qwen.py | da8479a |
| 2023-12-04 | update modeling_qwen.py | 9cae423 |
| 2023-12-03 | update modeling_qwen.py | ffc5701 |
| 2023-11-30 | update | 7f62181 |
| 2023-11-14 | remove fix-sized causal mask | 2abd8e5 |
| 2023-11-14 | update wechat.png | 0ca1b4c |
| 2023-11-05 | add kernel file check in modeling_qwen.py | 03752a6 |
| 2023-10-26 | update modeling.py | 013d71a |
| 2023-10-17 | update int8 quantization info | 2468096 |
| 2023-10-16 | update modeling_qwen.py | 8857f77 |
| 2023-10-14 | update batch inference | 33ae36b |
| 2023-10-13 | update default generate hyperparams | e1935c6 |
| 2023-10-12 | update react_prompt.md | 871c071 |
| 2023-10-11 | format | 7a6e27d |
| 2023-10-09 | update tokenization.py | de5bfb7 |
| 2023-10-09 | update tokenization.py | 4f49d06 |
| 2023-10-08 | Update README.md | 05d0727 |
| 2023-10-08 | Update README.md | f077bf1 |
| 2023-10-08 | Update README.md (#44) | df4768d |
| 2023-09-29 | Update README.md | 779ebe3 |
| 2023-09-28 | softmax_in_fp32 | faf3ff6 |
| 2023-09-27 | update modeling_qwen.py | 119ea93 |
| 2023-09-27 | update kernels | df4593d |
| 2023-09-26 | update modeling_qwen.py | 7df18a4 |
| 2023-09-25 | update modeling_qwen.py | 21ed8f8 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:38 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:08 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
