Up to date
Qwen Apache 2.0 out 18 months ago
1.72B parameters, text in, text out, 32K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #236 in LLMs · score 86.7 method v1 Sources 3 Downloads 30d 573.5k · #156 of 619 tracked
Changed this week: 2 new quantized builds on the Hub (51 in all).2026-10-24
Picked by rule from 19 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 4 picks above, plus these 15. 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; heavily run locally (51 quantized builds); a big base for builders (561 fine-tunes, adapters and merges).
Top Spaces running it: dqy08/InfoLens1 ♥ 22 · Ex0bit/qwen-scope-live ♥ 20 · Jellyfish042/Compression-Lens ♥ 4 · GiacomoDeLuca/orrery-demo ♥ 3
Weights only, an estimate: leave room for the KV cache. Method
About 1.0 GB at INT4, 2.1 GB at FP8, 4.1 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 4 dated articles (each links to it), grey ticks are repo commits.
612 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: 1,720,574,976 |
| Size band | S |
| 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 Qwen3-1.7B-Base: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 4.1 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 2.1 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 1.0 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
finetune: 478quantized: 51adapter: 79merge: 4
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| PrimeIntellect/Qwen3-1.7B | finetune | — | 2.03B | M | apache-2.0 | PrimeIntellect | 0 | 247.6k |
| unsloth/Qwen3-1.7B-Base | finetune | — | 1.72B | S | apache-2.0 | unsloth | 8 | 158k |
| unsloth/Qwen3-1.7B-Base-unsloth-bnb-4bit | quantized | -bnb | 1.76B | S | apache-2.0 | unsloth | 3 | 106.5k |
| OpenMOSS-Team/MOSS-TTSD-v0.5 | finetune | — | 2.05B | M | apache-2.0 | OpenMOSS-Team | 54 | 93k |
| occ-ai/OCC-RAG-1.7B | finetune | — | 1.72B | S | mit | occ-ai | 16 | 33.7k |
| unsloth/Qwen3-1.7B-Base-bnb-4bit | quantized | -bnb | 1.76B | S | apache-2.0 | unsloth | 0 | 27.9k |
| miulab/Qwen3-1.7B-Usefulness | finetune | — | 1.72B | S | — | miulab | 1 | 23.8k |
| jinaai/jina-vlm | finetune | — | 2.48B | M | cc-by-nc-4.0 | jinaai | 125 | 16.1k |
| CSHaitao/LegalOne-4B | finetune | — | 4.41B | M | mit | CSHaitao | 5 | 18.2k |
| QuantFactory/Qwen3-1.7B-GGUF | quantized | GGUF | — | — | apache-2.0 | QuantFactory | 3 | 12.2k |
| Qwen/Qwen3-1.7B-MLX-4bit | quantized | MLX | 1.72B | S | apache-2.0 | Qwen | 11 | 11.5k |
| Qwen/Qwen3-1.7B-MLX-8bit | quantized | MLX | 1.72B | S | apache-2.0 | Qwen | 9 | 10.4k |
| mradermacher/Qwen3-1.7B-Base-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 1 | 10.7k |
| Scicom-intl/Multilingual-TTS-1.7B-Base | finetune | — | 1.85B | S | — | Scicom-intl | 6 | 9k |
| OpenMOSS-Team/MOSS-TTSD-v0.7 | finetune | — | 2.05B | M | apache-2.0 | OpenMOSS-Team | 18 | 8.3k |
| Thrillcrazyer/Qwen3_1.7B_LoopUS | finetune | — | 2.03B | M | apache-2.0 | Thrillcrazyer | 3 | 6.2k |
| Qwen/Qwen3-1.7B-MLX-6bit | quantized | MLX | 1.72B | S | apache-2.0 | Qwen | 8 | 5.8k |
| Aratako/MioTTS-1.7B | finetune | — | 1.75B | S | apache-2.0 | Aratako | 14 | 5.4k |
| Polygl0t/Tucano2-qwen-1.5B-Base | finetune | — | 1.51B | S | apache-2.0 | Polygl0t | 3 | 5.8k |
| VLSP2025-LegalSML/qwen3-1.7b-legal-pretrain | finetune | — | 1.72B | S | — | VLSP2025-LegalSML | 0 | 5.8k |
| CSHaitao/LegalOne-8B | finetune | — | 8.20B | M | mit | CSHaitao | 9 | 5.2k |
| Qwen/Qwen3-1.7B-MLX-bf16 | finetune | MLX | 1.72B | S | apache-2.0 | Qwen | 12 | 4.9k |
| Thinking-Space/Qwen3-1.7B-SFT | finetune | — | 2.03B | M | other | Thinking-Space | 4 | 5.2k |
| malaysia-ai/Qwen3-1.7B-Multilingual-TTS | finetune | — | 1.85B | S | — | malaysia-ai | 11 | 4.4k |
| mradermacher/Qwen3-1.7B-Base-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 1 | 4.5k |
| eulogik/Bharat-Tiny-LLM-v3 | quantized | GGUF | 1.72B | S | apache-2.0 | eulogik | 6 | 4k |
| Mungert/Qwen3-1.7B-GGUF | quantized | GGUF | — | — | apache-2.0 | Mungert | 0 | 4k |
| Qwen/SAE-Res-Qwen3-1.7B-Base-W32K-L0_50 | finetune | — | — | — | other | Qwen | 14 | 2.6k |
| sds-ai/Yee-R1-mini-GGUF | quantized | GGUF | — | — | apache-2.0 | sds-ai | 4 | 3.1k |
| Abner0803/Qwen3-1.7B-msmarco-text-100k-with_pseudo_queries | finetune | — | 1.72B | S | — | Abner0803 | 0 | 3.2k |
| heyalexchoi/qwen3-1.7b-math-grpo | finetune | — | 1.72B | S | — | heyalexchoi | 0 | 3.1k |
| ofloxacin/qwen3-1b | finetune | — | 2.03B | M | apache-2.0 | ofloxacin | 2 | 2.8k |
| OpenMOSS-Team/MOSS-TTSD-v0 | finetune | — | 2.05B | M | apache-2.0 | OpenMOSS-Team | 28 | 1.3k |
| bluelightai/clt-qwen3-1.7b-base-20k | finetune | — | — | — | apache-2.0 | bluelightai | 1 | 2.5k |
| chartreuse-verte/prose-rewriter-1.7b-v1.2 | quantized | GGUF | 2.03B | M | agpl-3.0 | chartreuse-verte | 0 | 2.4k |
| YixuanEvenXu/Qwen3-1.7B-Base-HIP-adapter | adapter | — | — | — | apache-2.0 | YixuanEvenXu | 0 | 2.3k |
| ericoh929/qwen3-1.7b-lamini-qlora-instruction-tuned | adapter | — | 1.72B | S | cc-by-nc-4.0 | ericoh929 | 5 | 1.9k |
| Qwen/SAE-Res-Qwen3-1.7B-Base-W32K-L0_100 | finetune | — | — | — | other | Qwen | 9 | 1.7k |
| DevQuasar/Qwen.Qwen3-1.7B-Base-GGUF | quantized | GGUF | — | — | — | DevQuasar | 0 | 2k |
| ayushshah/Qwen3-1.7B-UltraChat-SFT | finetune | — | 1.72B | S | apache-2.0 | ayushshah | 1 | 1.9k |
| Date | Commit | SHA |
|---|---|---|
| 2025-07-26 | Create LICENSE | ea980cb |
| 2025-05-21 | Update README.md | 2944ab5 |
| 2025-05-20 | update tokenizer_config.json | 912d272 |
| 2025-04-28 | Update README.md | a0fbe33 |
| 2025-04-28 | Update tokenizer_config.json | 36be17a |
| 2025-04-28 | Upload merges.txt | 8b1d62d |
| 2025-04-28 | Update README.md | 4e9b56a |
| 2025-04-28 | Upload folder using huggingface_hub | 658b6ee |
| 2025-04-28 | initial commit | db5b20d |
Hosted endpoints observed for Qwen3-1.7B-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/Qwen3-1.7B-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.
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 21:37 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:05 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.
