HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Qwen3-0.6B

Qwen/Qwen3-0.6B

kind
original
quant
bf16 (16-bit)
family
qwen
modality
text-generation
root model
Qwen/Qwen3-0.6B-Base
provider
Qwen
license
apache-2.0
language
—
context
32,768
library
transformers
vRAM (GB)
—
size on disk
—
downloads
28,062,952
likes
1,453
created
—
last modified
2025-07-26T03:46:27+00:00 (2025)
huggingface
https://huggingface.co/Qwen/Qwen3-0.6B
tags
transformerssafetensorsqwen3text-generationconversationalarxiv:2505.09388base_model:Qwen/Qwen3-0.6B-Basebase_model:finetune:Qwen/Qwen3-0.6B-Baselicense:apache-2.0text-generation-inferenceendpoints_compatibledeploy:azureregion:usdeploy:sagemaker

Benchmark scores (0)

no scores extracted for this variant.

Discussions (38)

titleauthorstatuscommentstype
Create qwen3_0.6BPengFeiGaoopen0disc
Multilingual powerhouse — testing for mobile deployment3morixdopen0disc
Incredible speed on mobile devices3morixdopen0disc
Fix spelling mistake on model card!MihaiPopa-1open0disc
New architecture: TemporalMesh Transformer — dynamic kNN graph attention + per-token exit routing, 29.4 PPL at 48% computevigneshwar234open0disc
Where can I find the source code?work19990909open0disc
remove --enable-reasoning as it was deprecatedsaidmukhamadopen0disc
CopyrightDaveid151open0disc
Wasn't able to produce beanchmaked results for standard datasetsGroot18closed0disc
🚩 Report: Copyright infringementtomProfileclosed0disc
what is the max length token of qwen3-0.6Bguruboy123open0disc
benzionzion122draft0disc
qwen3_pull_from_huggingfacenah20250424draft0disc
BFloat16 is not supported on MPSTo-the-north-poleopen0disc
Memory spikes on long context lengthimomayizopen0disc
fgrrtkingnitdraft0disc
MMLU Redux Zero-shot Promptsh0416open0disc
qwen3zhai147draft0disc
Why 0.6B?yukiarimoclosed0disc
Config does not match context length.aberrioopen0disc
Request: DOIHaiOiopen0disc
Qwen3_0.6B is not fully compatible with the Qualcomm QCS6490 chipset. How can it be adapted? Can you provide a detailed and complete deployment tutorial?ThinkStartopen0disc
Thinking flag is flipped?rasbtopen0disc
How to use "user-defined output dimensions"?BigPan98closed0disc
Add assistant mask support to Qwen3-0.6Bwalekoopen0disc
Qwen3ForCausalLMcr-gknopen0disc
What's the difference between Qwen/Qwen3-0.6B and Qwen/Qwen3-0.6B-Base?luohuashijieyoufengjunopen0disc
Request: DOIAlexZrclosed0disc
Adding ONNX file of this modelMayuliopen0disc
Adding ONNX file of this modelMayuliclosed0disc
Adding ONNX file of this modelMayuliclosed0disc
Create aaaLinsonssopen0disc
Qwen3 0.6B model Bad Cases Genuinue User Reviews and Comments CollectionDeepNLPopen0disc
Very coherent for the size, though it is also adamant in defending its wrong conclusions and false knowledgecmp-nctopen0disc
Training methodologysLeroyDyeropen0disc
Update README.mdCAPsMANyoclosed0disc
【Evaluation】Best practice for evaluating Qwen3 !!wangxingjun778open0disc
Add languages tagde-francophonesopen0disc
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.