Model Follower.

Up to date

Rebuilt
2026-10-25 06:54 UTC
Release radar
117 on watch
Hub snapshot
2026-10-25
Re-checked
daily · every number sourced and dated

Models › Qwen3 family

Qwen3-4B

Hugging Face ↗ Follow changes

Qwen Apache 2.0 out 18 months ago

4.02B parameters, text in, text out, 40K tokens of context. 1 provider serves it.

Verified 2026-10-25 Rank #64 in LLMs · score 104.8 method v1 Sources 3 Downloads 30d 11.4M · #8 of 619 tracked

Changed this week: 14 new quantized builds on the Hub (359 in all).2026-10-24

Can I run it?2.4GB at INT4Estimatefits a 16 GB cardmethod
Can I use it?Apache 2.0Permissive; patent grantlicence file2026-10-24
Do people use it?11.4Mdownloads, 30 days#8 of 619 we track · top 1%+8 likes, downloads −3% this week770 likesHugging Face2026-10-24

Worth reading

Picked by rule from 24 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Most discussed · 490 HN pts

GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2

OpenAI released gpt-oss-120b and gpt-oss-20b, their first open-weight models since GPT-2 in 2019.

  • A 2025 research paper confirms Dropout results in worse downstream performance in single-epoch regimes.
magazine.sebastianraschka.com 2025-08-10 490 HN pts · 97 comments 5.2k words about the Qwen3 family family

24 best of 24 tracked

The 3 picks above, plus these 21. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

  1. ArticleQwen3-Coder-Nextqwen.ai · 2026-02-03 · official
  2. ArticleQwen3-Max-Thinkingqwen.ai · 2026-01-26 · official
  3. ArticleQwen3-TTS family is now open sourced: Voice design, clone, and generationqwen.ai · 2026-01-22 · official
  4. ArticleQwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computersventurebeat.com · 2026-02-28 · 461 HN pts
  5. ArticleQwen3: Think deeper, act fasterqwenlm.github.io · 2025-04-28 · 869 HN pts · official
  6. ArticleQwen3-Coder: Agentic coding in the worldqwenlm.github.io · 2025-07-22 · official
  7. ArticleQwen3-Nextqwen.ai · 2025-09-12 · official
  8. ArticleQwen3-VLqwen.ai · 2025-09-23 · official
  9. ArticleGood Results when training Qwen 3 4B to learn a new domainteachmecoolstuff.com · 2026-08-21 · 5 HN pts
  10. CodeQwen3-Omni: Native Omni AI model for text, image and videogithub.com · 2025-09-22
  11. ArticleQwen3 4B - API Pricing & Benchmarks | OpenRouteropenrouter.ai · undated
  12. ArticleQwen3-4B-Thinking: "This is art–pelicans don't ride bikes "simonwillison.net · 2025-08-14 · 4 HN pts
  13. CodeLatest progress helping Qwen3-4B Learngithub.com · 2026-03-01 · 2 HN pts
  14. ArticleQwen3 Deep Dive: Comprehensive Analysis of Alibaba’s Open-Source LLM Family and Its Variants – FTS AI & Roboticsflechetechstream.com · 2026-07-28
  15. ArticleTeaching Qwen3-4B to Trade: From Hold-Collapse to +9.4% Returnssabareesh.com · 2026-03-13
  16. Article[Literature Review] Qwen3 Technical Reportthemoonlight.io · undated
  17. ArticleQwen3 4B — 4B Dense LLM | FitMyLLMfitmyllm.com · undated
  18. ArticleQwen3 4B — pricing, benchmarks and overview 2026 · nAIvigatenaivigate.com · undated
  19. ArticleFine‑tuning Qwen3‑4B on a MacBook M1 for customer supportblog.layerx.xyz · 2026-04-17
  20. ArticleQwen3 Review: Hybrid Thinking Modes and MoE Architecture Explained — Efflooweffloow.com · 2026-04-13
  21. ArticleQwen3-VL-4B and 8B runs locally on NPU, GPU, and CPU with one SDKnexa.ai · 2025-10-14

Where to run it

Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.

#Cheapest live rowIn /1MOut /1MContextQuantSeen
1 Featherless — — — — Oct 25

All 1 provider →

Where it's used

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.

Served by 1 API provider; heavily run locally (359 quantized builds); a big base for builders (2,419 fine-tunes, adapters and merges).

  • As a service1Featherless
  • On own hardware359quantized builds
  • Built on2,4191,190 fine-tunes · 1,209 adapters · 20 merges
  • In apps100+Hugging Face Spaces

Top Spaces running it: Qwen/Qwen3-Demo ♥ 868 · pliny-the-prompter/obliteratus ♥ 621 · mrfakename/yue2-3b ♥ 221 · embedl/hfviewer ♥ 181

Will it fit?

Weights only, an estimate: leave room for the KV cache. Method

Fits a 16 GB card at INT4

About 2.4 GB at INT4, 4.8 GB at FP8, 9.7 GB at BF16: weights only, so leave room for the KV cache.

16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 2.4 GB of weights (estimate; KV cache and activations extra)INT42.4 GBFP8: about 4.8 GB of weights (estimate; KV cache and activations extra)FP84.8 GBBF16: about 9.7 GB of weights (estimate; KV cache and activations extra)BF169.7 GB
16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 2.4 GB of weights (estimate; KV cache and activations extra)INT42.4 GBFP8: about 4.8 GB of weights (estimate; KV cache and activations extra)FP84.8 GBBF16: about 9.7 GB of weights (estimate; KV cache and activations extra)BF169.7 GB

Estimate params × bytes per param × 1.2 · how we estimate

The story so far

Weights, every dated article, and repo commits on one rail.

9 dated articles and 19 repo commits so far

Teal is the weights (2025-04-27), amber triangles are 9 dated articles (each links to it), grey ticks are repo commits.

MayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOct2025-07-26: 1 repo commit2025-05-21: 1 repo commit2025-05-19: 1 repo commit2025-04-29: 1 repo commit2025-04-28: 14 repo commits2025-04-27: 1 repo commitQwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers · 2026-02-28Qwen3: Think deeper, act faster · 2025-04-28Show HN: An open-weight LLM whose answer/stop decision can be flipped internally · 2026-09-09GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2 · 2025-08-10Good Results when training Qwen 3 4B to learn a new domain · 2026-08-21Qwen3-4B-Thinking: "This is art–pelicans don't ride bikes " · 2025-08-14Latest progress helping Qwen3-4B Learn · 2026-03-01Qwen3 Deep Dive: Comprehensive Analysis of Alibaba’s Open-Source LLM Family and Its Variants – FTS AI & Robotics · 2026-07-28I swept all 667 layer duplication configs for Qwen3-4B on 1x RTX 3090 · 2026-04-14Weights 2025-04-27 · 2025-04-27Weights 2025-04-27now · 1 provider
MayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOct2025-07-26: 1 repo commit2025-05-21: 1 repo commit2025-05-19: 1 repo commit2025-04-29: 1 repo commit2025-04-28: 14 repo commits2025-04-27: 1 repo commitQwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers · 2026-02-28Qwen3: Think deeper, act faster · 2025-04-28Show HN: An open-weight LLM whose answer/stop decision can be flipped internally · 2026-09-09GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2 · 2025-08-10Good Results when training Qwen 3 4B to learn a new domain · 2026-08-21Qwen3-4B-Thinking: "This is art–pelicans don't ride bikes " · 2025-08-14Latest progress helping Qwen3-4B Learn · 2026-03-01Qwen3 Deep Dive: Comprehensive Analysis of Alibaba’s Open-Source LLM Family and Its Variants – FTS AI & Robotics · 2026-07-28I swept all 667 layer duplication configs for Qwen3-4B on 1x RTX 3090 · 2026-04-14Weights 2025-04-27 · 2025-04-27Weights 2025-04-27now · 1 provider

What people built on it

2,778 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

2,778 derivatives on the Hub, led by 1,209 adapters

From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.

359 quantized (Hugging Face model tree)1,190 fine-tunes (Hugging Face model tree)20 merges (Hugging Face model tree)1,209 adapters (Hugging Face model tree)359 quantized1,190 fine-tunes20 merges1,209 adapters
359 quantized (Hugging Face model tree)1,190 fine-tunes (Hugging Face model tree)20 merges (Hugging Face model tree)1,209 adapters (Hugging Face model tree)359 quantized1,190 fine-tunes20 merges1,209 adapters
GGUF MaziyarPanahi/Qwen3-4B-GGUF for llama.cpp, Ollama, LM Studio 3.6M downloads MLX lmstudio-community/Qwen3-4B-MLX-4bit for Apple silicon 432.6k downloads AWQ Qwen/Qwen3-4B-AWQ for vLLM, SGLang 4.1M downloads

All variants →

Reference

Everything else, folded: the full tables, every source.

What the card says

1 cited sentence

Summary generated, cited

The model architecture uses 36 layers and supports a native context length of 32,768 tokens.[1][2]

  1. “Number of Layers: 36”
  2. “Context Length: 32,768 natively”

Generated 2026-10-07 15:00 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.

Use-case evidence

6 use cases

What it is good at 6 use cases

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.

Facts

21 fields

Listed in LLMs

Facts

Pipeline tagtext-generation
Librarytransformers
ArchitecturesQwen3ForCausalLM
Model typeqwen3
Dtypebfloat16
Params by dtypeBF16: 4,022,468,096
Size bandM
MoEno
Context length40,960 (config.json max_position_embeddings)
PapersarXiv:2309.00071, arXiv:2505.09388
LicenseApache 2.0 (apache-2.0)
Hub snapshot2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: supported · Qwen3ForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Qwen3ForCausalLM · 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.

Hardware fit

3 estimates

Hardware fit ESTIMATE

Weights-only memory for Qwen3-4B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)9.7 GB · Single GPU (<24 GB)
FP8 (1 byte/param)4.8 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)2.4 GB · Edge (<4 GB)

Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.

All variants

40 variants

Lineage

Fine-tuned from Qwen/Qwen3-4B-Base (same organization) Family: Qwen/Qwen3 2,778 derivatives on the Hub

Base model: Qwen/Qwen3-4B-Base

quantized: 359adapter: 1,209finetune: 1,190merge: 20

Top variants (40 shown, ranked by downloads)

Relation
Quant
Size
40 rows
Variant Relation Quant Params Band License Author Likes Downloads
Qwen/Qwen3-4B-AWQ quantized AWQ 4.02B M apache-2.0 Qwen 38 4.1M
MaziyarPanahi/Qwen3-4B-GGUF quantized GGUF — — — MaziyarPanahi 9 3.6M
diff-interpretation-tuning/loras adapter — — — mit diff-interpretation-tuning 2 3.3M
Qwen/Qwen3-4B-GGUF quantized GGUF — — apache-2.0 Qwen 204 2.5M
unsloth/Qwen3-4B-GGUF quantized GGUF — — apache-2.0 unsloth 252 2.4M
zeroentropy/zerank-2-reranker finetune — 4.02B M apache-2.0 zeroentropy 145 2.2M
unsloth/Qwen3-4B finetune — 4.02B M — unsloth 20 2.2M
unsloth/Qwen3-4B-unsloth-bnb-4bit quantized -bnb 4.02B M apache-2.0 unsloth 22 1.6M
YannQi/R-4B finetune — 4.82B M apache-2.0 YannQi 188 1.3M
Qwen/Qwen3-4B-FP8 quantized -FP8 4.41B M apache-2.0 Qwen 46 1.1M
AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-GGUF quantized GGUF — — apache-2.0 AnkitAI 26 1.1M
zeroentropy/zembed-1-embedding finetune — 4.02B M apache-2.0 zeroentropy 133 572.5k
unsloth/Qwen3-4B-bnb-4bit quantized -bnb 4.14B M — unsloth 3 547k
lmstudio-community/Qwen3-4B-MLX-4bit quantized MLX 4.02B M apache-2.0 lmstudio-community 1 432.6k
lmstudio-community/Qwen3-4B-MLX-8bit quantized MLX 4.02B M apache-2.0 lmstudio-community 0 412.9k
bartowski/Qwen_Qwen3-4B-GGUF quantized GGUF — — — bartowski 22 332.8k
mlx-community/Qwen3-4B-4bit quantized -4bit 4.02B M apache-2.0 mlx-community 17 251.3k
lmstudio-community/Qwen3-4B-GGUF quantized GGUF — — — lmstudio-community 20 208.5k
zeroentropy/zerank-1-small-reranker finetune — 1.72B S apache-2.0 zeroentropy 67 194.9k
Qwen/Qwen3-4B-SafeRL finetune — 4.02B M apache-2.0 Qwen 52 178.7k
mlxha/Qwen3-4B-grpo-medmcqa finetune — 4.02B M — mlxha 2 175k
farbodtavakkoli/OTel-Reranker-4B finetune — 4.02B M apache-2.0 farbodtavakkoli 0 168.1k
RedHatAI/Qwen3-4B-quantized.w4a16 quantized -quant 4.44B M apache-2.0 RedHatAI 4 130k
Skywork/Skywork-Reward-V2-Qwen3-4B finetune — 4.02B M apache-2.0 Skywork 11 118.4k
Menlo/Jan-nano finetune — 4.02B M apache-2.0 Menlo 509 75k
mlc-ai/Qwen3-4B-q4f16_1-MLC quantized — — — — mlc-ai 2 95.1k
geoffmunn/Qwen3-4B-f16 quantized GGUF — — apache-2.0 geoffmunn 6 86.4k
unsloth/Qwen3-4B-128K-GGUF quantized GGUF — — apache-2.0 unsloth 27 68.8k
arianraje/qwen3-4b-mamba3-hybrid-stage2b-kd-bias0 finetune — 4.74B M apache-2.0 arianraje 1 62.8k
Klingspor/Qwen3-4B-SFT finetune — 4.41B M apache-2.0 Klingspor 1 60.5k
electricpipelines/Qwen3-4B-GGUF quantized GGUF — — apache-2.0 electricpipelines 0 53.5k
ggml-org/Qwen3-4B-GGUF quantized GGUF — — apache-2.0 ggml-org 8 53k
zeroentropy/zerank-1-reranker finetune — 4.02B M apache-2.0 zeroentropy 81 47.9k
pytorch/Qwen3-4B-INT8-INT4 quantized INT8 — — mit pytorch 2 47.6k
DreamFast/qwen3-4b-heretic quantized GGUF 4.02B M apache-2.0 DreamFast 43 43.8k
litert-community/Qwen3-4B quantized — — — apache-2.0 litert-community 11 39.3k
mradermacher/Qwen3-4B-i1-GGUF quantized GGUF — — apache-2.0 mradermacher 5 39k
mlx-community/Qwen3-4B-8bit quantized -8bit 4.02B M apache-2.0 mlx-community 1 32.4k
huihui-ai/Huihui-Qwen3-4B-abliterated-v2 finetune — 4.02B M apache-2.0 huihui-ai 66 25.4k
SeerAttention/SeerAttention-Decode-Qwen3-4B-AttnGates adapter — — — mit SeerAttention 2 25.7k

Repository changelog

19 commits
changed in the last 7 daysOther: 2026-10-24all changes

Repository changelog 19 commits

DateCommitSHA
2025-07-26 Create LICENSE 1cfa9a7
2025-05-21 Update README.md 531c80e
2025-05-19 update tokenizer_config.json 8136a03
2025-04-29 Update README.md 82d62bb
2025-04-28 Update README.md 9e1b55c
2025-04-28 Update README.md 76ae75a
2025-04-28 Update README.md 05960ad
2025-04-28 Delete special_tokens_map.json 7873580
2025-04-28 Delete added_tokens.json 396eb12
2025-04-28 Update README.md 2863ddf
2025-04-28 Update README.md 3fb3566
2025-04-28 Update README.md ff64285
2025-04-28 Update README.md a4e9bba
2025-04-28 Update README.md e619e22
2025-04-28 Update README.md eb971e9
2025-04-28 Update README.md 7c69a10
2025-04-28 Update README.md 350135a
2025-04-28 Upload folder using huggingface_hub 3101254
2025-04-27 initial commit 8af1666

All providers

1 provider · 1 endpoint

Where to run 1 providers · 1 endpoints

Hosted endpoints observed for Qwen3-4B, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.

ProviderModel refContextQuant$/M in$/M outtok/sTTFTToolsStatusObserved
Featherless
hf
 Qwen/Qwen3-4B — — — — — — — live 2026-10-25

Articles and coverage

24 pieces

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.

Official1

Qwen3: Think deeper, act faster
qwenlm.github.io2025-04-28officialauthorityHN front pageabout the Qwen3 familyHN 869 pts · 388 comments2,139 words
Qwen3 introduces Hybrid Thinking Modes, allowing users to toggle between step-by-step reasoning and rapid responses via computational budget control. The open-weight MoE model Qwen3-235B-A22B achieves performance similar to dense models while using only 10% of the active parameters. Pre-training utilized approximately 36 trillion tokens across 119 languages, with deployment supported by SGLang and vLLM.written, reviewed, cited
  • Qwen3-8B and Qwen3-14B models do not use tied embeddings, unlike smaller variants.
  • The pre-training pipeline includes a final stage extending context length to 32K tokens.

Analysis9

Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers
venturebeat.com2026-02-28HN front pageabout the Qwen3 familyHN 461 pts · 272 comments
Show HN: An open-weight LLM whose answer/stop decision can be flipped internally
github.com2026-09-09authorityHN 1 pts · 0 comments193 words
GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2
magazine.sebastianraschka.com2025-08-10authorityHN front pageabout the Qwen3 familyHN 490 pts · 97 comments5,195 words
OpenAI released gpt-oss-120b and gpt-oss-20b, their first open-weight models since GPT-2 in 2019. The 20B model can run on a consumer GPU with up to 16 GB of RAM, and the 120B model can run on a single H100 with 80 GB of RAM or newer hardware. The models utilize MXFP4 optimization to fit onto single GPUs.written, reviewed, cited
  • A 2025 research paper confirms Dropout results in worse downstream performance in single-epoch regimes.
Good Results when training Qwen 3 4B to learn a new domain
teachmecoolstuff.com2026-08-21HN 5 pts · 0 comments
Qwen3 4B - API Pricing & Benchmarks | OpenRouter
openrouter.aiauthority
Qwen3-4B-Thinking: "This is art–pelicans don't ride bikes "
simonwillison.net2025-08-14authorityHN 4 pts · 0 comments
Latest progress helping Qwen3-4B Learn
github.com2026-03-01authorityHN 2 pts · 0 comments
Qwen3 Deep Dive: Comprehensive Analysis of Alibaba’s Open-Source LLM Family and Its Variants – FTS AI & Robotics
flechetechstream.com2026-07-28about the Qwen3 family
I swept all 667 layer duplication configs for Qwen3-4B on 1x RTX 3090
austinsnerdythings.com2026-04-14HN 3 pts · 0 comments
All coverage tracked for this model (24) — append-only, nothing is ever removed
TitleSourcePublishedFirst seenSignals
Qwen3 4B - API Pricing & Benchmarks | OpenRouteropenrouter.ai—2026-10-07exa
[Literature Review] Qwen3 Technical Reportthemoonlight.io—2026-10-07exa
Qwen3 4B — 4B Dense LLM | FitMyLLMfitmyllm.com—2026-10-07exa
Qwen3 4B — pricing, benchmarks and overview 2026 · nAIvigatenaivigate.com—2026-10-07exa
Show HN: An open-weight LLM whose answer/stop decision can be flipped internallygithub.com2026-09-092026-10-07hn
Good Results when training Qwen 3 4B to learn a new domainteachmecoolstuff.com2026-08-212026-10-07hn
Qwen3 Deep Dive: Comprehensive Analysis of Alibaba’s Open-Source LLM Family and Its Variants – FTS AI & Roboticsflechetechstream.com2026-07-282026-10-07exa
Fine‑tuning Qwen3‑4B on a MacBook M1 for customer supportblog.layerx.xyz2026-04-172026-10-07hn
I swept all 667 layer duplication configs for Qwen3-4B on 1x RTX 3090austinsnerdythings.com2026-04-142026-10-07hn
Qwen3 Review: Hybrid Thinking Modes and MoE Architecture Explained — Efflooweffloow.com2026-04-132026-10-07exa
Teaching Qwen3-4B to Trade: From Hold-Collapse to +9.4% Returnssabareesh.com2026-03-132026-10-07hn
Latest progress helping Qwen3-4B Learngithub.com2026-03-012026-10-07hn
Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computersventurebeat.com2026-02-282026-10-07hn
Qwen3-Coder-Nextqwen.ai official2026-02-032026-10-07hn
Qwen3-Max-Thinkingqwen.ai official2026-01-262026-10-07hn
Qwen3-TTS family is now open sourced: Voice design, clone, and generationqwen.ai official2026-01-222026-10-07hn
Qwen3-VL-4B and 8B runs locally on NPU, GPU, and CPU with one SDKnexa.ai2025-10-142026-10-07hn
Qwen3-VLqwen.ai official2025-09-232026-10-07hn
Qwen3-Omni: Native Omni AI model for text, image and videogithub.com2025-09-222026-10-07hn
Qwen3-Nextqwen.ai official2025-09-122026-10-07hn
Qwen3-4B-Thinking: "This is art–pelicans don't ride bikes "simonwillison.net2025-08-142026-10-07hn
GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2magazine.sebastianraschka.com2025-08-102026-10-07hn
Qwen3-Coder: Agentic coding in the worldqwenlm.github.io official2025-07-222026-10-07hn
Qwen3: Think deeper, act fasterqwenlm.github.io official2025-04-282026-10-07hn

Papers

2 papers

Papers 2

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).

Similar models

3 models

Similar models

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:33 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:00 UTC · Last verified 2026-10-25

Follow Qwen3-4B

Watch this model

Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.