Up to date
Qwen Apache 2.0 out 18 months ago
14.8B parameters, text in, text out, 32K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #411 in LLMs · score 79.2 method v1 Sources 3 Downloads 30d 100.4k · #314 of 619 tracked
Picked by rule from 10 tracked pieces: the lab's own post, the most shared, a field report, the latest.
OpenAI released gpt-oss-120b and gpt-oss-20b, their first open-weight models since GPT-2 in 2019.
The 2 picks above, plus these 8. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | Featherless | — | — | — | — | Oct 25 |
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; 28 quantized builds for local use; a big base for builders (153 fine-tunes, adapters and merges).
Top Spaces running it: dqy08/InfoLens1 ♥ 22 · Girlz/TokenTrace ♥ 1
Weights only, an estimate: leave room for the KV cache. Method
About 8.9 GB at INT4, 18 GB at FP8, 35 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 (2025-04-28), amber triangles are 3 dated articles (each links to it), grey ticks are repo commits.
181 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 | Qwen3ForCausalLM |
| Model type | qwen3 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 14,768,307,200 |
| Size band | L |
| MoE | no |
| Context length | 32,768 (config.json max_position_embeddings) |
| Papers | arXiv:2505.09388 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
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.
Weights-only memory for Qwen3-14B-Base: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 35 GB · Single high-end GPU (<80 GB) |
|---|---|
| FP8 (1 byte/param) | 18 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 8.9 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
finetune: 107quantized: 28merge: 4adapter: 42
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| OpenPipe/Qwen3-14B-Instruct | finetune | — | 14.8B | L | apache-2.0 | OpenPipe | 15 | 2.2M |
| unsloth/Qwen3-14B-Base-unsloth-bnb-4bit | quantized | -bnb | 14.8B | L | apache-2.0 | unsloth | 5 | 98.3k |
| AlphaGaO/Qwen3-14B-GPTQ | quantized | GPTQ | 14.8B | L | apache-2.0 | AlphaGaO | 1 | 45.6k |
| unsloth/Qwen3-14B-Base | finetune | — | 14.8B | L | apache-2.0 | unsloth | 2 | 29.1k |
| Qwen/Qwen3-14B-MLX-8bit | quantized | MLX | 14.8B | L | apache-2.0 | Qwen | 13 | 27k |
| McGill-NLP/AfriqueQwen-14B | finetune | — | 14.8B | L | cc-by-4.0 | McGill-NLP | 4 | 20.3k |
| Qwen/Qwen3-14B-MLX-4bit | quantized | MLX | 14.8B | L | apache-2.0 | Qwen | 26 | 17.2k |
| winglian/qwen3-14b-math | finetune | — | 14.8B | L | apache-2.0 | winglian | 0 | 16.9k |
| DevQuasar/Qwen.Qwen3-14B-Base-GGUF | quantized | GGUF | — | — | — | DevQuasar | 0 | 16.2k |
| Cylingo/Xinyuan-LLM-14B-0428 | finetune | — | 14.8B | L | apache-2.0 | Cylingo | 16 | 11.6k |
| unsloth/Qwen3-14B-Base-bnb-4bit | quantized | -bnb | 14.8B | L | apache-2.0 | unsloth | 0 | 10.7k |
| Mungert/Qwen3-14B-GGUF | quantized | GGUF | — | — | apache-2.0 | Mungert | 8 | 8.2k |
| snuh/hari-q3-14b | finetune | — | 14.8B | L | apache-2.0 | snuh | 12 | 7.9k |
| mradermacher/Qwen3-14B-Base-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 2 | 7.4k |
| TIGER-Lab/General-Reasoner-Qwen3-14B | finetune | — | 14.8B | L | apache-2.0 | TIGER-Lab | 6 | 5.6k |
| HelpingAI/Dhanishtha-2.0-preview | finetune | — | 14.8B | L | apache-2.0 | HelpingAI | 58 | 2.9k |
| mradermacher/Qwen3-14B-Base-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 2 | 5.3k |
| Qwen/Qwen3-14B-MLX-6bit | quantized | MLX | 14.8B | L | apache-2.0 | Qwen | 9 | 4.6k |
| MegaScience/Qwen3-14B-MegaScience | finetune | — | 14.8B | L | apache-2.0 | MegaScience | 6 | 4.1k |
| hkust-nlp/drkernel-14b-coldstart | finetune | — | 462M | S | — | hkust-nlp | 0 | 3.5k |
| Qwen/Qwen3-14B-MLX-bf16 | finetune | MLX | 14.8B | L | apache-2.0 | Qwen | 10 | 2.6k |
| Userb1az/Qwen3-14B-GGUF | quantized | GGUF | — | — | apache-2.0 | Userb1az | 0 | 2.7k |
| hkust-nlp/drkernel-14b | finetune | — | 14.8B | L | — | hkust-nlp | 6 | 2.4k |
| 1nstaller/Qwen3-14B-NoThinking | finetune | — | 14.8B | L | apache-2.0 | 1nstaller | 0 | 2.5k |
| togethercomputer/HybridAgent-14B-Agentic | finetune | — | 14.8B | L | mit | togethercomputer | 17 | 1.3k |
| abeja/ABEJA-Qwen3-14B-Agentic-256k-v0.1 | finetune | — | 14.8B | L | apache-2.0 | abeja | 8 | 1.2k |
| AscendKernelGen/KernelGen-LM-14B | finetune | — | 14.8B | L | apache-2.0 | AscendKernelGen | 1 | 1.5k |
| avi686/Qwen3-14B-heretic | quantized | GGUF | 14.8B | L | apache-2.0 | avi686 | 2 | 1.4k |
| Antigma/Qwen3-14B-Base-GGUF | quantized | GGUF | — | — | apache-2.0 | Antigma | 0 | 1.5k |
| Azure99/Blossom-V6.3-14B-GGUF | quantized | GGUF | — | — | apache-2.0 | Azure99 | 0 | 1.4k |
| liodon-ai/Qwen3-14B-Base-imatrix-GGUF | quantized | imatrix | — | — | other | liodon-ai | 1 | 1k |
| xing720310/qwen3-14b | finetune | — | 14.8B | L | apache-2.0 | xing720310 | 0 | 917 |
| ForSureTesterSim/Qwen3-14B-NSFW-FDA | finetune | — | 14.8B | L | apache-2.0 | ForSureTesterSim | 2 | 816 |
| ForSureTesterSim/Big-Randy-NSFW-Beta-14B | merge | — | 14.8B | L | — | ForSureTesterSim | 0 | 720 |
| aimeri/spoomplesmaxx-mini-14B | finetune | — | 14.8B | L | apache-2.0 | aimeri | 1 | 652 |
| itlwas/Qwen3-14B-Base-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | itlwas | 0 | 674 |
| Darth-Coder/my-model3-14b-it | finetune | — | 14.8B | L | apache-2.0 | Darth-Coder | 0 | 644 |
| Open4bits/Qwen3-14B-Base-GGUF | quantized | GGUF | — | — | — | Open4bits | 0 | 639 |
| noahclark556/mesh-to-cad-14b | quantized | — | 14.8B | L | cc-by-nc-4.0 | noahclark556 | 3 | 474 |
| KKHYA/llavaqwen3-14b-finetune | finetune | — | 15.1B | L | apache-2.0 | KKHYA | 0 | 522 |
| Date | Commit | SHA |
|---|---|---|
| 2025-07-26 | Create LICENSE | 0b0bd37 |
| 2025-05-21 | Update README.md | 8aef508 |
| 2025-05-20 | update tokenizer_config.json | e5faa17 |
| 2025-04-28 | Update README.md | ae3040d |
| 2025-04-28 | Update tokenizer_config.json | b22103a |
| 2025-04-28 | Upload merges.txt | eb64b25 |
| 2025-04-28 | Update README.md | de5dcbc |
| 2025-04-28 | Upload folder using huggingface_hub | ec06e99 |
| 2025-04-28 | initial commit | 4ad9a24 |
Hosted endpoints observed for Qwen3-14B-Base, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| Featherless hf | Qwen/Qwen3-14B-Base | — | — | — | — | — | — | — | live | 2026-10-25 |
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.
| Title | Source | Published | First seen | Signals |
|---|---|---|---|---|
| Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers | venturebeat.com | 2026-02-28 | 2026-10-07 | hn |
| Qwen3-Coder-Next | qwen.ai official | 2026-02-03 | 2026-10-07 | hn |
| Qwen3-Max-Thinking | qwen.ai official | 2026-01-26 | 2026-10-07 | hn |
| Qwen3-TTS family is now open sourced: Voice design, clone, and generation | qwen.ai official | 2026-01-22 | 2026-10-07 | hn |
| Qwen3-VL | qwen.ai official | 2025-09-23 | 2026-10-07 | hn |
| Qwen3-Omni: Native Omni AI model for text, image and video | github.com | 2025-09-22 | 2026-10-07 | hn |
| Qwen3-Next | qwen.ai official | 2025-09-12 | 2026-10-07 | hn |
| GPT-OSS vs. Qwen3 and a detailed look how things evolved since GPT-2 | magazine.sebastianraschka.com | 2025-08-10 | 2026-10-07 | hn |
| Qwen3-Coder: Agentic coding in the world | qwenlm.github.io official | 2025-07-22 | 2026-10-07 | hn |
| Qwen3: Think deeper, act faster | qwenlm.github.io official | 2025-04-28 | 2026-10-07 | hn |
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 21:39 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:10 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.
