DeepSeek R1 Distill Qwen 14B

DeepSeek R1 Distill Qwen 14B

DeepSeek · Released Jan 20, 2025 Emerging
Intelligence #506 / 583
14.9 Our Score
AA Index #228 / 380
9.8 Artificial Analysis
Input
— Not priced
Output
— Not priced
Context
— Not reported

Analysis Summary

DeepSeek R1 Distill Qwen 14B is a distilled version of DeepSeek's R1 reasoning model, built on a 14B Qwen backbone. With an intelligence index of 9.8, strong GPQA performance (0.484), and a LiveCodeBench score of 0.376, it punches above its weight for a model of this size, particularly on math and science reasoning tasks.

For businesses, it is best suited to self-hosted deployments where cost control is critical and tasks involve structured reasoning, math, or code generation. Instruction-following scores are modest (IFBench 0.221) and long-context reliability is limited (LCR 0.07), so it is not well-suited to complex multi-step agentic workflows or long-document analysis.

As an open-weight model, it offers flexibility for teams with infrastructure to run it locally. For cloud-based business use, newer and more capable models in the DeepSeek family or from Mistral offer better overall performance.

Assessed June 17, 2026

Editorial notes

DeepSeek R1 Distill Qwen 14B is a distilled reasoning model with strong math and GPQA scores, capable on coding tasks, and available as an open-weight model for self-hosted deployment.

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence2.3Technical0Value0Content5
Intelligence 2.3/10
Technical 0/10
Content 5/10
Value 0/10

How DeepSeek R1 Distill Qwen 14B compares

DeepSeek R1 Distill Qwen 14B ranks #228 of 380 AI models we track for overall intelligence. DeepSeek R1 Distill Qwen 14B is currently free to use via OpenRouter.

Performance Indices

Source: Artificial Analysis

9.8 Intelligence Index
55.7 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 48.4% Graduate-level scientific reasoning
HLE 4.4% Humanity's Last Exam
MMLU Pro 74% Multi-task language understanding
MATH 500 94.9% Mathematical problem-solving
AIME 66.7% Competition mathematics
AIME 2025 55.7% Competition mathematics (2025)
SciCode 23.9% Scientific computing

Technical

LiveCodeBench 37.6% Live coding evaluation

Content

IFBench 22.1% Instruction following
LCR 7% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does DeepSeek R1 Distill Qwen 14B stack up?

Compare side-by-side with other emerging models.

Compare Models

Model Information

ProviderDeepSeek
Release Date January 20, 2025
Status Active

Leaderboard Categories

Frequently asked questions about DeepSeek R1 Distill Qwen 14B

How much does DeepSeek R1 Distill Qwen 14B cost?

DeepSeek R1 Distill Qwen 14B is currently available for free via OpenRouter.

Who created DeepSeek R1 Distill Qwen 14B?

DeepSeek R1 Distill Qwen 14B is developed by DeepSeek and was released on January 20, 2025.