Qwen2-0.5B

494M parameters, text in, text out, 128K tokens of context. 1 provider serves it.

Verified 2026-10-26 07:27 UTC Rank #251 in LLMs · score 85.9 method v1 Sources 3 Downloads 30d 211.5k · #243 of 619 tracked

Can I run it?0.3GB at INT4Estimatefits a 16 GB cardmethod

Can I use it?Apache 2.0Permissive; patent grantlicence file2026-10-26

Do people use it?211.5kdownloads, 30 days#243 of 619 we track · top 39%+3 likes, downloads −18% this week180 likesHugging Face2026-10-26

Worth reading

Picked by rule from 11 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Most discussed · 29 HN pts

Alibaba Qwen2.5-Omni-7B: Open Source End-to-End Multimodal AI Model

alizila.com 2025-03-27 29 HN pts · 3 comments about the Qwen2 family family

Latest · 21 months ago

Qwen2.5-Max: The Next Leap in AI Innovation

medium.com 2025-01-29 17 HN pts · 4 comments about the Qwen2 family family

11 best of 11 tracked

The 4 picks above, plus these 7. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

ArticleQwen2 LLM Releasedqwenlm.github.io · 2024-06-06 · 261 HN pts · official
ArticleQwen2-Mathqwenlm.github.io · 2024-08-08 · 128 HN pts · official
CodeLlama.cpp Now Supports Qwen2-VL (Vision Language Model)github.com · 2024-12-14
ArticleAn Analysis of Chinese LLM Censorship and Bias with Qwen 2 Instructhuggingface.co · 2024-06-09 · 16 HN pts
ArticleQwen2 Technical Report — Paper Graphpaper-graph-ui.vercel.app · undated
ArticleQwen2-0.5B: Lightweight Open-Source LLM for Edgedev.co · undated
ArticleQwen2 0.5B Instruct · Models · Dataloopdataloop.ai · undated

Show all 7

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 26

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; 42 quantized builds for local use; a big base for builders (710 fine-tunes, adapters and merges).

As a service1Featherless
On own hardware42quantized builds
Built on710146 fine-tunes · 562 adapters · 2 merges
In apps50Hugging Face Spaces

Top Spaces running it: microsoft/MInference ♥ 61 · linxianzhong0128/Linly-Talker ♥ 5 · xcczach/mini-omni-s2s ♥ 3 · Thouph/prompt2tag-qwen2-0.5b-v0.1-demo ♥ 2

Will it fit?

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

Fits a 16 GB card at INT4

About 0.3 GB at INT4, 0.6 GB at FP8, 1.2 GB at BF16: weights only, so leave room for the KV cache.

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

The story so far

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

5 dated articles and 6 repo commits so far

Teal is the weights (2024-05-31), amber triangles are 5 dated articles (each links to it), grey ticks are repo commits.

What people built on it

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

752 derivatives on the Hub, most of them adapters

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

GGUF legraphista/Qwen2-0.5B-IMat-GGUF for llama.cpp, Ollama, LM Studio 54.7k downloads MLX mlx-community/dolphin-2.9.3-qwen2-0.5b-4bit for Apple silicon 1.6k downloads

All variants →

Reference

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

Benchmarks

14 rows

Benchmarks 14 rows

measured third-party leaderboard rowauthor-reported number from the model card or eval files (14)No 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
author-reported ARC Challenge · Chat and follow instructions	31.5	—	model author	2026-10-07
author-reported BBH · Reason through hard problems	28.4	—	model author	2026-10-07
author-reported C-Eval · Work across languages	58.2	—	model author	2026-10-07
author-reported CMMLU · Work across languages	55.1	—	model author	2026-10-07
author-reported GSM8K · Reason through hard problems	36.5	—	model author	2026-10-07
author-reported HellaSwag · Chat and follow instructions	49.3	—	model author	2026-10-07
author-reported HumanEval · Write code with agents	22	—	model author	2026-10-07
author-reported MATH-500 · Reason through hard problems	10.7	—	model author	2026-10-07
author-reported MBPP · Write code with agents	22	—	model author	2026-10-07
author-reported MMLU · Reason through hard problems	45.4	—	model author	2026-10-07
author-reported MMLU-Pro · Reason through hard problems	14.7	—	model author	2026-10-07
author-reported Theorem QA	8.9	—	model author	2026-10-07
author-reported TruthfulQA · Chat and follow instructions	39.7	—	model author	2026-10-07
author-reported Winogrande · Chat and follow instructions	56.8	—	model author	2026-10-07

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

Qwen2-0.5B has 494M parameters. (safetensors metadata, observed 2026-10-26)
Context length is 131,072 tokens (128K). (config.json max_position_embeddings, observed 2026-10-26)
License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-26)
Created 2024-05-31, last modified 2024-10-22. (HF API, observed 2026-10-26)

Pipeline tag	text-generation
Library	transformers
Architectures	Qwen2ForCausalLM
Model type	qwen2
Dtype	bfloat16
Params by dtype	BF16: 494,032,768
Size band	S
MoE	no
Context length	131,072 (config.json max_position_embeddings)
Languages	en
License	Apache 2.0 (apache-2.0)
Hub snapshot	2026-10-26

Servability

vLLM · SGLang

Servability

vLLM: supported · Qwen2ForCausalLM · 4b1c04e76fe9ac3ca8b7e541dd423e7d5f38dc60SGLang: supported · Qwen2ForCausalLM · 60d941293e3af27a12c6eb64e2c3e2e1a17f100f

vLLM registry (4b1c04e76fe9ac3ca8b7e541dd423e7d5f38dc60) · SGLang registry (60d941293e3af27a12c6eb64e2c3e2e1a17f100f) · 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 Qwen2-0.5B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)	1.2 GB · Edge (<4 GB)
FP8 (1 byte/param)	0.6 GB · Edge (<4 GB)
INT4 (0.5 bytes/param)	0.3 GB · Edge (<4 GB)

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

All variants

40 variants

Lineage

Original model Family: Qwen/Qwen2 752 derivatives on the Hub

quantized: 42finetune: 146adapter: 562merge: 2

Top variants (40 shown, ranked by downloads)

Variant	Relation	Quant	Params	Band	License	Author	Likes	Downloads
legraphista/Qwen2-0.5B-IMat-GGUF	quantized	GGUF	—	—	apache-2.0	legraphista	0	54.7k
QuantFactory/Qwen2-0.5B-GGUF	quantized	GGUF	—	—	apache-2.0	QuantFactory	3	36.3k
gpt-omni/mini-omni	finetune	—	—	—	mit	gpt-omni	445	6
HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v2	finetune	—	494M	S	mit	HIT-TMG	34	13.9k
mradermacher/Qwen2-0.5B-GGUF	quantized	GGUF	—	—	apache-2.0	mradermacher	1	10.1k
mradermacher/Qwen2-0.5B-i1-GGUF	quantized	GGUF	—	—	apache-2.0	mradermacher	0	6.4k
hZzy/qwen2-0.5b-sft	finetune	—	494M	S	apache-2.0	hZzy	0	5.5k
dphn/dolphin-2.9.3-qwen2-0.5b	finetune	—	494M	S	apache-2.0	dphn	6	4.5k
second-state/Qwen2-0.5B-GGUF	quantized	GGUF	—	—	apache-2.0	second-state	1	4.1k
afrideva/Qwen2-0.5B-GGUF	quantized	GGUF	—	—	apache-2.0	afrideva	1	3.6k
JiachenFu/Qwen2-0.5B-detectanyllm-detector-en	adapter	—	—	—	—	JiachenFu	1	2.7k
Kyleyee/Qwen2-0.5B-DPO-imdb-bm	finetune	—	494M	S	—	Kyleyee	0	2.6k
DevQuasar-8/Qwen.Qwen2-0.5B-GGUF	quantized	GGUF	—	—	—	DevQuasar-8	0	2.4k
Kyleyee/Qwen2-0.5B-DRDPO-imdb-bm-tp	finetune	—	494M	S	—	Kyleyee	0	2.1k
Kyleyee/Qwen2-0.5B-DRDPO-imdb-bm-rp	finetune	—	494M	S	—	Kyleyee	0	2.1k
Kyleyee/Qwen2-0.5B-DRDPO-imdb-bm-wp	finetune	—	494M	S	—	Kyleyee	0	2k
mlx-community/dolphin-2.9.3-qwen2-0.5b-4bit	finetune	-4bit	77.3M	S	apache-2.0	mlx-community	3	1.6k
FreedomIntelligence/Apollo-MoE-0.5B	finetune	—	787M	S	apache-2.0	FreedomIntelligence	3	1.5k
trollek/Qwen2-0.5B-DiffusionPrompter-v0.1-GGUF	quantized	GGUF	—	—	apache-2.0	trollek	5	1.2k
FreedomIntelligence/Apollo2-0.5B	finetune	—	630M	S	apache-2.0	FreedomIntelligence	1	1.2k
ZzWater/viitor-voice-mix	finetune	—	637M	S	cc-by-nc-sa-4.0	ZzWater	5	826
protecttors/sample-files	quantized	GGUF	—	—	apache-2.0	protecttors	0	1.1k
Romarchive/Qwen2-0.5B-GGUF	quantized	GGUF	—	—	apache-2.0	Romarchive	0	1.1k
liodon-ai/Qwen2-0.5B-imatrix-GGUF	quantized	imatrix	—	—	other	liodon-ai	0	1k
Kyleyee/Qwen2-0.5B-stf-imdb	finetune	—	494M	S	—	Kyleyee	0	1k
bnm8901/humanizerAI_0.5	finetune	—	494M	S	apache-2.0	bnm8901	0	941
trl-lib/Qwen2-0.5B-Reward-Math-Sheperd	finetune	—	494M	S	—	trl-lib	1	888
Kyleyee/Qwen2-0.5B-stf-hh	finetune	—	494M	S	—	Kyleyee	0	855
CoolSpring/Qwen2-0.5B-Abyme	finetune	—	494M	S	apache-2.0	CoolSpring	0	827
TsipiDev/Thesis_Models_Dimitris_Vatousis_220007	quantized	GGUF	—	—	other	TsipiDev	0	817
JiachenFu/Qwen2-0.5B-detectanyllm-detector-zh	adapter	—	—	—	—	JiachenFu	0	806
amy011872/LawToken-0.5B-a2	finetune	—	505M	S	apache-2.0	amy011872	0	795
ksanjeeb/PipaT1-500M	quantized	GGUF	—	—	apache-2.0	ksanjeeb	2	597
123-cao/Qwen2-0.5B-Instruct	finetune	—	494M	S	apache-2.0	123-cao	0	693
trollek/Qwen2-0.5B-DiffusionPrompter-v0.1	finetune	—	630M	S	apache-2.0	trollek	2	580
dolly-vn/Vira-TTS	finetune	—	507M	S	apache-2.0	dolly-vn	4	479
ljwclass/Qwen2-0.5B-Instruct	finetune	—	494M	S	apache-2.0	ljwclass	0	677
musk12/apple-fastvlm-qwen2-q4-gguf	quantized	gguf	—	—	apache-2.0	musk12	0	568
mlx-community/dolphin-2.9.3-qwen2-0.5b-8bit	finetune	-8bit	139M	S	apache-2.0	mlx-community	2	446
leafspark/mini-omni-safetensors	finetune	—	694M	S	mit	leafspark	5	283

Repository changelog

6 commits

Repository changelog 6 commits

Date	Commit	SHA
2024-10-22	add metadata for newer model version (#7)	91d2aff
2024-06-06	update README.md	ff3a49f
2024-06-06	Update README.md	3b832a9
2024-05-31	Update config.json	ce45c39
2024-05-31	Initial commit	b6aca0b
2024-05-31	initial commit	01c2382

All providers

1 provider · 1 endpoint

Where to run 1 providers · 1 endpoints

Hosted endpoints observed for Qwen2-0.5B, 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/Qwen2-0.5B	—	—	—	—	—	—	—	live	2026-10-26

Articles and coverage

11 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.

Official3

Qwen2 LLM Released

qwenlm.github.io2024-06-06officialauthorityHN front pageabout the Qwen2 familyHN 261 pts · 130 comments4,713 words

Alibaba releases Qwen2, a new series of open-source models featuring five sizes ranging from 0.5B to 72B parameters. The update applies Group Query Attention to all sizes and trains on data in 27 additional languages besides English and Chinese. Specific instruction-tuned versions like Qwen2-7B-Instruct and Qwen2-72B-Instruct support extended context lengths up to 128K tokens, while smaller models adopt the Apache 2.0 license and the largest retains the Qianwen License.written, reviewed, cited

Qwen2-57B-A14B disables tie embedding because large sparse embeddings consume a large proportion of total model parameters.
Instruction-tuned models require the Needle in a Haystack task to verify correct long-context understanding beyond perplexity evaluation.
The Qwen2-72B-Instruct model performs comparably to GPT-4 in safety tests across four categories of multilingual unsafe queries.

Qwen2-Math

qwenlm.github.io2024-08-08officialauthorityHN front pageabout the Qwen2 familyHN 128 pts · 38 comments5,833 words

Qwen2-Math is a series of specialized math language models built upon the Qwen2 LLMs that significantly outperforms the mathematical capabilities of open-source models and even closed-source models. The largest model, Qwen2-Math-72B-Instruct, is evaluated on benchmarks including GSM8K, Math, and MMLU-STEM using few-shot chain-of-thought prompting. Instruction tuning combines a dense reward signal with a binary signal via Rejection Sampling and reinforcement learning with Group Relative Policy Optimization (GRPO) after SFT.written, reviewed, cited

The model series primarily supports English, with bilingual English and Chinese math models scheduled for release soon.
Evaluation of multi-choice benchmarks including MMLU STEM and GaoKao problems uses a 5-shot setting instead of zero-shot.
Performance reporting for complex competition evaluations includes Greedy, Maj@64, RM@64, and RM@256 metrics.

An Analysis of Chinese LLM Censorship and Bias with Qwen 2 Instruct

huggingface.co2024-06-09authorityabout the Qwen2 familyHN 16 pts · 6 comments3,368 words

Qwen 2 Instruct models operate under an Apache 2.0 license but exhibit unique government-mandated alignment with significantly (>80%) fewer refusals in Chinese than English. The deccp dataset documents these behaviors, showing that responses to sensitive topics often include defensive nationalist propaganda or warnings rather than standard refusals. While the models are competitive on benchmarks, their output varies wildly in tone depending on the language used.written, reviewed, cited

Qwen2-72B-Instruct ranks just below Llama-3-70b-Instruct on the ChatArena Leaderboard several days after release.
The deccp dataset contains approximately 95 refusal questions used to evaluate Qwen2 Instructs censorship behavior.

Analysis7

Alibaba Qwen2.5-Omni-7B: Open Source End-to-End Multimodal AI Model

alizila.com2025-03-27about the Qwen2 familyHN 29 pts · 3 comments

Qwen2.5-Max: The Next Leap in AI Innovation

medium.com2025-01-29about the Qwen2 familyHN 17 pts · 4 comments

Qwen2 Technical Report — Paper Graph

paper-graph-ui.vercel.appabout the Qwen2 family

Qwen2 0.5B · Benchmarks, Pricing & Performance | BenchGecko

benchgecko.ai193 words

Qwen2-0.5B: Lightweight Open-Source LLM for Edge

dev.co932 words

Qwen2 0.5B Instruct · Models · Dataloop

dataloop.ai

Qwen2-0.5B vs Qwen2.5-0.5B-Instruct: Specs, License and GPU Cost

savrn.com

All coverage tracked for this model (11) — append-only, nothing is ever removed

Title	Source	Published	First seen	Signals
Qwen2 Technical Report — Paper Graph	paper-graph-ui.vercel.app	—	2026-10-07	exa
Qwen2 0.5B · Benchmarks, Pricing & Performance \| BenchGecko	benchgecko.ai	—	2026-10-07	exa
Qwen2-0.5B: Lightweight Open-Source LLM for Edge	dev.co	—	2026-10-07	exa
Qwen2 0.5B Instruct · Models · Dataloop	dataloop.ai	—	2026-10-07	exa
Qwen2-0.5B vs Qwen2.5-0.5B-Instruct: Specs, License and GPU Cost	savrn.com	—	2026-10-07	exa
Alibaba Qwen2.5-Omni-7B: Open Source End-to-End Multimodal AI Model	alizila.com	2025-03-27	2026-10-07	hn
Qwen2.5-Max: The Next Leap in AI Innovation	medium.com	2025-01-29	2026-10-07	hn
Llama.cpp Now Supports Qwen2-VL (Vision Language Model)	github.com	2024-12-14	2026-10-07	hn
Qwen2-Math	qwenlm.github.io official	2024-08-08	2026-10-07	hn
An Analysis of Chinese LLM Censorship and Bias with Qwen 2 Instruct	huggingface.co	2024-06-09	2026-10-07	hn
Qwen2 LLM Released	qwenlm.github.io official	2024-06-06	2026-10-07	hn

Similar models

3 models

Similar models

Qwen2.5-0.5B-Instruct · 494M
same size band (S), same task (text-generation), permissive license
Qwen2.5-0.5B · 494M
same size band (S), same task (text-generation), permissive license
Qwen2.5-Coder-0.5B-Instruct · 494M
same size band (S), same task (text-generation), permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-26 · HF API: 2026-10-26 07:19 UTC · Providers: 2026-10-26 07:26 UTC · Articles: 2026-10-26 07:27 UTC · Summary: 2026-10-07 15:05 UTC · Last verified 2026-10-26 07:27 UTC

Follow Qwen2-0.5B

Watch this model

/feeds/models/Qwen--Qwen2-0.5B.xml(RSS 2.0) · every change to Qwen2-0.5B
/feeds/families/Qwen--Qwen2.xml(RSS 2.0) · every change across the Qwen2 family

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