Qwen3-1.7B

2.03B parameters, text in, text out, 40K tokens of context. 1 provider serves it.

Verified 2026-10-25 Rank #82 in LLMs · score 102.9 method v1 Sources 3 Downloads 30d 4.1M · #35 of 619 tracked

Changed this week: 18 new quantized builds on the Hub (435 in all).2026-10-24 · 17 new quantized builds on the Hub (434 in all).2026-10-24

Can I run it?1.2GB at INT4Estimatefits a 16 GB cardmethodCommunity: Show HN: App running Qwen3-1.7b_q4f16 in browser featuring Bender from Futurama →

Can I use it?Apache 2.0Permissive; patent grantlicence file2026-10-24

Does it work?28.41on BFCL#21 of 28 measuredleaderboard2026-10-25

Do people use it?4.1Mdownloads, 30 days#35 of 619 we track · top 6%+10 likes, downloads +2% this week592 likesHugging Face2026-10-24

Worth reading

Picked by rule from 23 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

Latest · 8 months ago

Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers

venturebeat.com 2026-02-28 461 HN pts · 272 comments about the Qwen3 family family

23 best of 23 tracked

The 4 picks above, plus these 19. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

ArticleQwen3-Coder-Nextqwen.ai · 2026-02-03 · official
ArticleQwen3-Max-Thinkingqwen.ai · 2026-01-26 · official
ArticleQwen3-TTS family is now open sourced: Voice design, clone, and generationqwen.ai · 2026-01-22 · official
ArticleQwen3: Think deeper, act fasterqwenlm.github.io · 2025-04-28 · 869 HN pts · official
ArticleQwen3-Coder: Agentic coding in the worldqwenlm.github.io · 2025-07-22 · official
ArticleQwen3-Nextqwen.ai · 2025-09-12 · official
ArticleQwen3-VLqwen.ai · 2025-09-23 · official
CodeQwen3-Omni: Native Omni AI model for text, image and videogithub.com · 2025-09-22
PaperQwen3 Technical Report - arXivarxiv.org · undated
ArticleQwen3 1.7B (non-reasoning) - Artificial Analysisartificialanalysis.ai · undated
ArticleUnderstanding and Implementing Qwen3 From Scratchmagazine.sebastianraschka.com · undated
ArticleQwen3 1.7B (Reasoning) vs Qwen3 4B (Reasoning): Model Comparisonartificialanalysis.ai · undated
ArticleQwen3 1.7B Fine-Tuning Guide – distil labsdistillabs.ai · 2026-08-02
CodeC inference for Qwen3-ASR 0.6B and 1.7B transcriptions modelsgithub.com · 2026-02-28
DiscussionQwen 3 1.7B running locally on mobile at 30 tokens/stwitter.com · 2025-05-16
CodeShow HN: Qwen3-TTS integration for Manim animationsgithub.com · 2026-01-24
CodeShow HN: App running Qwen3-1.7b_q4f16 in browser featuring Bender from Futuramagithub.com · 2025-09-13
ArticleQwen3 1.7B | Open Laboratoryopenlaboratory.com · undated
ArticleShow HN: Using Qwen3:1.7B to call itself recursivelyseanneilan.com · 2026-01-16

Show all 19

Where to run it

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

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 (435 quantized builds); a big base for builders (2,135 fine-tunes, adapters and merges).

As a service1Featherless
On own hardware435quantized builds
Built on2,1351,329 fine-tunes · 800 adapters · 6 merges
In apps100+Hugging Face Spaces

Top Spaces running it: Qwen/Qwen3-Demo ♥ 868 · pliny-the-prompter/obliteratus ♥ 621 · embedl/hfviewer ♥ 181 · tomg-group-umd/lm-watermarking ♥ 100

Will it fit?

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

Fits a 16 GB card at INT4

About 1.2 GB at INT4, 2.4 GB at FP8, 4.9 GB at BF16: weights only, so leave room for the KV cache.

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

How it scores

Third-party measurements only; author-reported numbers stay in Reference, marked as such.

#21 of 28 on BFCL so far

28.41. Each tick is a model on this site with a measured BFCL row; author-reported numbers stay in Reference. Percent (overall accuracy). FC

Source · leaderboard · 2026-10-25

The story so far

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

4 dated articles and 18 repo commits so far

Teal is the weights (2025-04-27), amber triangles are 4 dated articles (each links to it), grey ticks are repo commits.

What people built on it

2,570 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

2,570 derivatives on the Hub, most of them fine-tunes

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

GGUF MaziyarPanahi/Qwen3-1.7B-GGUF for llama.cpp, Ollama, LM Studio 3.5M downloads MLX lmstudio-community/Qwen3-1.7B-MLX-8bit for Apple silicon 502.7k downloads GPTQ Qwen/Qwen3-1.7B-GPTQ-Int8 for vLLM, SGLang 1.5M downloads

All variants →

Reference

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

Benchmarks

1 row

Benchmarks 1 row

measured third-party leaderboard row (1)author-reported number from the model card or eval filesNo confidence intervals unless the source gave them; settings differ between rows, so compare within a benchmark, not across. Date = the benchmark's own date when it gives one, else when we read it.

Benchmark	Value	Setting / notes	Reported by	Date
measured BFCL · Call tools and APIs	28.41 percent (overall accuracy)	FC 5 sub-scores Non-Live AST Acc: 82.92 Live Acc: 74.61 Multi Turn Acc: 11 Web Search Acc: 2.5 Memory Acc: 6.02	leaderboard	2026-10-25

What the card says

1 cited sentence

Summary generated, cited

The model architecture features 28 layers and a context length of 32,768.^[1]^[2]

“Number of Layers: 28”
“Context Length: 32,768”

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

5 use cases

What it is good at 5 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

Qwen3-1.7B has 2.03B parameters. (safetensors metadata, observed 2026-10-24)
Context length is 40,960 tokens (40K). (config.json max_position_embeddings, observed 2026-10-24)
License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
Created 2025-04-27, last modified 2025-07-26. (HF API, observed 2026-10-24)

Pipeline tag	text-generation
Library	transformers
Architectures	Qwen3ForCausalLM
Model type	qwen3
Dtype	bfloat16
Params by dtype	BF16: 2,031,739,904
Size band	M
MoE	no
Context length	40,960 (config.json max_position_embeddings)
Papers	arXiv:2505.09388
License	Apache 2.0 (apache-2.0)
Hub snapshot	2026-10-25

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 Qwen3-1.7B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)	4.9 GB · Single GPU (<24 GB)
FP8 (1 byte/param)	2.4 GB · Edge (<4 GB)
INT4 (0.5 bytes/param)	1.2 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/Qwen3-1.7B-Base (same organization) Family: Qwen/Qwen3 2,570 derivatives on the Hub

Base model: Qwen/Qwen3-1.7B-Base

quantized: 435finetune: 1,329merge: 6adapter: 800

Top variants (40 shown, ranked by downloads)

Variant	Relation	Quant	Params	Band	License	Author	Likes	Downloads
MaziyarPanahi/Qwen3-1.7B-GGUF	quantized	GGUF	—	—	—	MaziyarPanahi	11	3.5M
Qwen/Qwen3-1.7B-GPTQ-Int8	quantized	GPTQ	1.72B	S	apache-2.0	Qwen	14	1.5M
ricdomolm/mini-coder-1.7b	finetune	—	1.72B	S	mit	ricdomolm	8	1.2M
unsloth/Qwen3-1.7B-unsloth-bnb-4bit	quantized	-bnb	1.76B	S	apache-2.0	unsloth	12	976.1k
unsloth/Qwen3-1.7B-GGUF	quantized	GGUF	—	—	apache-2.0	unsloth	79	966.1k
nvidia/canary-qwen-2.5b	finetune	—	2.56B	M	cc-by-4.0	nvidia	472	933.4k
unsloth/Qwen3-1.7B	finetune	—	1.72B	S	—	unsloth	13	938.4k
farbodtavakkoli/OTel-LLM-1.7B-IT	finetune	—	—	—	apache-2.0	farbodtavakkoli	1	713.8k
Qwen/Qwen3-1.7B-GGUF	quantized	GGUF	—	—	apache-2.0	Qwen	84	573.6k
tobil/qmd-query-expansion-1.7B-gguf	quantized	gguf	—	—	mit	tobil	25	546.7k
lmstudio-community/Qwen3-1.7B-MLX-8bit	quantized	MLX	1.72B	S	apache-2.0	lmstudio-community	1	502.7k
lmstudio-community/Qwen3-1.7B-MLX-4bit	quantized	MLX	1.72B	S	apache-2.0	lmstudio-community	0	486.6k
lmstudio-community/Qwen3-1.7B-MLX-bf16	finetune	MLX	1.72B	S	apache-2.0	lmstudio-community	5	483.3k
Qwen/Qwen3-1.7B-FP8	quantized	-FP8	2.03B	M	apache-2.0	Qwen	43	469.1k
dnotitia/Smoothie-Qwen3-1.7B	finetune	—	1.72B	S	apache-2.0	dnotitia	2	363k
mlx-community/Qwen3-1.7B-4bit	quantized	-4bit	1.72B	S	apache-2.0	mlx-community	7	336.9k
ggml-org/Qwen3-1.7B-GGUF	quantized	GGUF	—	—	apache-2.0	ggml-org	14	254.3k
lmstudio-community/Qwen3-1.7B-GGUF	quantized	GGUF	—	—	—	lmstudio-community	8	223.7k
PerceptronAI/Isaac-0.1	finetune	—	2.57B	M	cc-by-nc-4.0	PerceptronAI	118	216.1k
Vikhrmodels/QVikhr-3-1.7B-Instruction-noreasoning	finetune	—	1.72B	S	apache-2.0	Vikhrmodels	12	192.6k
bartowski/Qwen_Qwen3-1.7B-GGUF	quantized	GGUF	—	—	—	bartowski	24	188.6k
AI-MO/Kimina-Prover-Distill-1.7B	finetune	—	2.03B	M	apache-2.0	AI-MO	7	108.4k
unsloth/Qwen3-1.7B-bnb-4bit	quantized	-bnb	1.76B	S	—	unsloth	5	83.2k
XXXXyu/Qwen3-1.7B-w2g64-gptq_v2	quantized	gptq	2.03B	M	apache-2.0	XXXXyu	0	76.9k
tomg-group-umd/DynaGuard-1.7B	finetune	—	2.03B	M	apache-2.0	tomg-group-umd	3	66.7k
Skywork/Skywork-Reward-V2-Qwen3-1.7B	finetune	—	1.72B	S	apache-2.0	Skywork	8	65.1k
geoffmunn/Qwen3-1.7B-f16	quantized	GGUF	—	—	apache-2.0	geoffmunn	4	61.2k
Klingspor/Qwen3-1.7B-SFT	finetune	—	2.03B	M	apache-2.0	Klingspor	2	49.8k
contextboxai/Qwen3-1.7B-FC	finetune	—	1.72B	S	apache-2.0	contextboxai	4	46.6k
Orion-zhen/Qwen3-1.7B-AWQ	quantized	AWQ	1.72B	S	gpl-3.0	Orion-zhen	1	39.9k
mlc-ai/Qwen3-1.7B-q4f16_1-MLC	quantized	—	—	—	—	mlc-ai	0	37.2k
RedHatAI/Qwen3-1.7B-quantized.w4a16	quantized	-quant	2.08B	M	apache-2.0	RedHatAI	3	37k
NCSOFT/VARCO-VISION-2.0-1.7B	finetune	—	2.12B	M	cc-by-nc-4.0	NCSOFT	25	31.8k
activeDap/Qwen3-1.7B_hh_harmful	finetune	—	1.72B	S	apache-2.0	activeDap	0	30.7k
guangyangnlp/Qwen3-1.7B-SFT-science-2e-5	finetune	—	2.03B	M	other	guangyangnlp	0	28.7k
RedHatAI/Qwen3-1.7B-FP8-dynamic	quantized	-FP8	2.03B	M	apache-2.0	RedHatAI	0	28.1k
litert-community/Qwen3-1.7B	quantized	—	—	—	apache-2.0	litert-community	8	23.7k
UnfilteredAI/DAN-Qwen3-1.7B	finetune	—	1.72B	S	apache-2.0	UnfilteredAI	49	21.1k
NCSOFT/VARCO-VISION-2.0-1.7B-OCR	finetune	—	2.12B	M	cc-by-nc-4.0	NCSOFT	29	20.9k
huihui-ai/Huihui-Qwen3-1.7B-abliterated-v2	finetune	—	1.72B	S	apache-2.0	huihui-ai	5	21.9k

Repository changelog

18 commits

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

Repository changelog 18 commits

Date	Commit	SHA
2025-07-26	Create LICENSE	70d244c
2025-05-21	Update README.md	0060bc5
2025-05-19	update tokenizer_config.json	0ecc654
2025-04-29	Update README.md	d3e2589
2025-04-28	Update README.md	bf6c947
2025-04-28	Update README.md	44c8955
2025-04-28	Update README.md	33c2236
2025-04-28	Update README.md	d725c74
2025-04-28	Delete special_tokens_map.json	b9352fb
2025-04-28	Delete added_tokens.json	7794bc7
2025-04-28	Update README.md	9cbdf16
2025-04-28	Update README.md	026ef89
2025-04-28	Update README.md	0fa42c2
2025-04-28	Update README.md	2fe16ea
2025-04-28	Update README.md	442b4a6
2025-04-28	Update README.md	3f1b00a
2025-04-28	Upload folder using huggingface_hub	8d4744f
2025-04-27	initial commit	c52c071

All providers

1 provider · 1 endpoint

Where to run 1 providers · 1 endpoints

Hosted endpoints observed for Qwen3-1.7B, 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	—	—	—	—	—	—	—	live	2026-10-25

Articles and coverage

23 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

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.

Qwen3 Technical Report - arXiv

arxiv.orgauthorityabout the Qwen3 family

Qwen3 1.7B - Intelligence, Performance & Price Analysis | Artificial Analysis

artificialanalysis.aiauthority

Qwen3 1.7B (non-reasoning) - Artificial Analysis

artificialanalysis.aiauthority

Understanding and Implementing Qwen3 From Scratch

magazine.sebastianraschka.comauthorityabout the Qwen3 family

Qwen3 1.7B: API Provider Performance Benchmarking & Price Analysis | Artificial Analysis

artificialanalysis.aiauthority

Qwen3 1.7B (Reasoning) vs Qwen3 4B (Reasoning): Model Comparison

artificialanalysis.aiauthority

Qwen3 1.7B Fine-Tuning Guide – distil labs

distillabs.ai2026-08-02

All coverage tracked for this model (23) — append-only, nothing is ever removed

Title	Source	Published	First seen	Signals
Qwen3 Technical Report - arXiv	arxiv.org	—	2026-10-07	exa
Qwen3 1.7B - Intelligence, Performance & Price Analysis \| Artificial Analysis	artificialanalysis.ai	—	2026-10-07	exa
Qwen3 1.7B (non-reasoning) - Artificial Analysis	artificialanalysis.ai	—	2026-10-07	exa
Understanding and Implementing Qwen3 From Scratch	magazine.sebastianraschka.com	—	2026-10-07	exa
Qwen3 1.7B: API Provider Performance Benchmarking & Price Analysis \| Artificial Analysis	artificialanalysis.ai	—	2026-10-07	exa
Qwen3 1.7B (Reasoning) vs Qwen3 4B (Reasoning): Model Comparison	artificialanalysis.ai	—	2026-10-07	exa
Qwen3 1.7B \| Open Laboratory	openlaboratory.com	—	2026-10-07	exa
Qwen3 1.7B Fine-Tuning Guide – distil labs	distillabs.ai	2026-08-02	2026-10-07	exa
Qwen3.5 122B and 35B models offer Sonnet 4.5 performance on local computers	venturebeat.com	2026-02-28	2026-10-07	hn
C inference for Qwen3-ASR 0.6B and 1.7B transcriptions models	github.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
Show HN: Qwen3-TTS integration for Manim animations	github.com	2026-01-24	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
Show HN: Using Qwen3:1.7B to call itself recursively	seanneilan.com	2026-01-16	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
Show HN: App running Qwen3-1.7b_q4f16 in browser featuring Bender from Futurama	github.com	2025-09-13	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
Qwen 3 1.7B running locally on mobile at 30 tokens/s	twitter.com	2025-05-16	2026-10-07	hn
Qwen3: Think deeper, act faster	qwenlm.github.io official	2025-04-28	2026-10-07	hn, exa

Papers

1 paper

Papers 1

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).

Qwen3 Technical Report this model

An Yang, Anfeng Li, Baosong Yang +57 · 2025-05-14 · PDF · HF paper page (349 upvotes)

Similar models

3 models

Similar models

DeepSeek-R1-Distill-Qwen-1.5B · 1.78B
same task (text-generation), permissive license
Qwen1.5-1.8B-Chat · 1.84B
same task (text-generation)
Qwen1.5-1.8B · 1.84B
same task (text-generation)

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 Qwen3-1.7B

Watch this model

/feeds/models/Qwen--Qwen3-1.7B.xml(RSS 2.0) · every change to Qwen3-1.7B
/feeds/families/Qwen--Qwen3.xml(RSS 2.0) · every change across the Qwen3 family

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