NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/R1 Distill Qwen 32B
DeepSeek logo

R1 Distill Qwen 32B

by DeepSeek · Released Jan 2025

Open Source
41.9
avg score
Rank #188
Compare
Better than 40% of all models
Context
33K tokens (~16 books)
Input $/1M
$0.29
Output $/1M
$0.29
Type
text
License
Open Source
Benchmarks
6 tested
Data as of
About

DeepSeek R1 Distill Qwen 32B is a distilled large language model based on Qwen 2.5 32B, using outputs from DeepSeek R1. It outperforms OpenAI's o1-mini across various benchmarks, achieving new...

Tested on 6 benchmarks · BenchGecko score 41.9. Top scores: IFEval (41.9%), MMLU-PRO (41.0%), BBH (HuggingFace) (17.1%).

Looking for similar performance at lower cost?
Llama 3 8B Instruct scores 41.8 (100% as good) at $0.14/1M input · 52% cheaper
Capabilities
reasoning
16.1
#148 globally
math
17.1
#241 globally
knowledge
22.8
#257 globally
general
17.1
#173 globally
language
41.9
#132 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.

16.1·
mathCompare math →
MATH Level 5

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

17.1·
knowledgeCompare knowledge →
MMLU-PRO

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

41.0·
GPQA

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

4.6·
Quick compare:Meta logovs Llama 3 8B InstructGoogle DeepMind logovs Gemini 1.5 Flash (May 2024)Google DeepMind logovs Gemma 2 2b It
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Similar Models
Meta logo
Llama 3 8B Instruct
Meta
41.8$0.14/1M
Google DeepMind logo
Gemini 1.5 Flash (May 2024)
Google DeepMind
41.7n/a
Google DeepMind logo
Gemma 2 2b It
Google DeepMind
41.6n/a
Links
Info
DeepSeekPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@DeepSeek
Source Code
GitHubHugging Face
BenchGecko API
deepseek-r1-distill-qwen-32b
Specifications
  • Typetext
  • Context33K tokens (~16 books)
  • ReleasedJan 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.001
Available On
DeepSeek logoDeepSeek$0.29
Categories
reasoningmathknowledgegenerallanguage
Learn More
transformeropen weightstokens
Share & Export
Tweet
Related Models
Meta logoLlama 3 8B Instruct
41.8
Google DeepMind logoGemini 1.5 Flash (May 2024)
41.7
Google DeepMind logoGemma 2 2b It
41.6
Meta logoLlama 3-70B
41.6
R1 Distill Qwen 32B is an open-source text AI model by DeepSeek, released in January 2025. It has an average benchmark score of 41.9. Context window: 33K tokens.

Related Models

Llama 3 8B Instruct · MetaGemini 1.5 Flash (May 2024) · Google DeepMindGemma 2 2b It · Google DeepMindLlama 3-70B · MetaClaude Opus 4.1 · Anthropic

Benchmarks

IFEvalMMLU-PROBBH (HuggingFace)MATH Level 5MUSR

Related Pages

DeepSeek · ProviderDeepSeek · EconomyAll ModelsCompare ModelsPricingDevelopers · API

Key facts · as of 2026-05-03

  • R1 Distill Qwen 32B by DeepSeek. BenchGecko score 41.9, rank 188 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.29 input · $0.29 output per 1M tokens (as of 2026-05-03).

How to cite · data as of 2026-05-03

R1 Distill Qwen 32B · benchmarks, pricing and providers. BenchGecko, data as of 2026-05-03. https://benchgecko.ai/model/deepseek-r1-distill-qwen-32b

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