NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Qwen2 0.5B
Alibaba logo

Qwen2 0.5B

by Alibaba · Released May 2024

Open Source
10.8
avg score
Rank #298
Compare
Better than 4% of all models
Context
N/A
Input $/1M
n/a
Output $/1M
n/a
Type
text-generation
License
Open Source
Benchmarks
6 tested
Data as of
About

Qwen text generation model. 497K downloads on HuggingFace.

Tested on 6 benchmarks · BenchGecko score 10.8. Top scores: IFEval (18.7%), MMLU-PRO (8.0%), BBH (HuggingFace) (7.9%).

Capabilities
reasoning
4.6
#205 globally
math
2.6
#270 globally
knowledge
4.7
#294 globally
general
7.9
#198 globally
language
18.7
#166 globally
Benchmark Scores
Compare All
Tested on 6 benchmarks · Ranked across 5 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
reasoningCompare reasoning →
MUSR

HuggingFace MuSR (Multi-Step Reasoning). Tests multi-hop reasoning requiring chaining multiple facts together.

4.6·
mathCompare math →
MATH Level 5

HuggingFace evaluation of MATH Level 5 problems. Competition math requiring advanced reasoning and proof construction.

2.6·
knowledgeCompare knowledge →
MMLU-PRO

HuggingFace MMLU-Pro. Harder version of MMLU with 10 answer choices instead of 4 and more challenging questions.

8.0·
GPQA

HuggingFace evaluation of GPQA (Graduate-Level Google-Proof Q&A). PhD-level science questions that cannot be easily searched.

1.4·
Quick compare:NVIDIA logovs Nemotron 3 SuperGoogle DeepMind logovs Gemma 3 12BMeta logovs Llama 3.2 1B Instruct
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Alibaba Qwen 2
May 2024
10.8
n/an/actx6 benchmarks
Jun 2024
7.8-3.0
n/an/actx6 benchmarks
Jun 2024
23.9+16.1
n/an/actx6 benchmarks
Jun 2024
76.5+52.6
n/an/actx25 benchmarks
Aug 2024
n/an/actx
Aug 2024
64.9-11.6
n/an/actx11 benchmarks
Aug 2024
n/an/actx
See the full Qwen 2 family →
Similar Models
NVIDIA logo
Nemotron 3 Super
NVIDIA
10.2$0.08/1M
Google DeepMind logo
Gemma 3 12B
Google DeepMind
10.2$0.05/1M
Meta logo
Llama 3.2 1B Instruct
Meta
11.7$0.03/1M
Links
Info
AlibabaPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Alibaba
Source Code
GitHubHugging Face
BenchGecko API
qwen-qwen2-05b
Specifications
  • Typetext-generation
  • ContextN/A
  • ReleasedMay 2024
  • LicenseOpen Source
  • StatusActive
Available On
Alibaba logoAlibaban/a
Categories
reasoningmathknowledgegenerallanguage
Learn More
transformeropen weightstokens
Share & Export
Tweet
Related Models
NVIDIA logoNemotron 3 Super
10.2
Google DeepMind logoGemma 3 12B
10.2
Meta logoLlama 3.2 1B Instruct
11.7
eleutherai logoPythia 160m
9.9
Qwen2 0.5B is an open-source text-generation AI model by Alibaba, released in May 2024. It has an average benchmark score of 10.8.

Related Models

Nemotron 3 Super · NVIDIAGemma 3 12B · Google DeepMindLlama 3.2 1B Instruct · MetaPythia 160m · eleutheraiGpt2 · OpenAI

Benchmarks

IFEvalMMLU-PROBBH (HuggingFace)MUSRMATH Level 5

Related Pages

Alibaba · ProviderAlibaba · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-04-09

  • Qwen2 0.5B by Alibaba. BenchGecko score 10.8, rank 298 of 312 scored models (normalized average of public benchmark scores).
  • List price n/a input · n/a output per 1M tokens (as of 2026-04-09).

How to cite · data as of 2026-04-09

Qwen2 0.5B · benchmarks, pricing and providers. BenchGecko, data as of 2026-04-09. https://benchgecko.ai/model/qwen-qwen2-05b

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP