How fast is DeepSeek R1 Distill Qwen 32B?

60 output tokens per second, 0.8 s to the first token.

Is DeepSeek R1 Distill Qwen 32B open source?

Yes. Its weights are published.

DeepSeek R1 Distill Qwen 32B

DeepSeek R1 Distill Qwen 32B is a reasoning model from DeepSeek in the DeepSeek R1 Distill family. It has 5 scored results in 4 categories. It has no index.

Open weights
Reasoning
Input: text
Released 2025-01

Unranked

IndexUnrankedMedian 51.0

Coverage45%

Speed60/sMedian 67/s

First token0.8 sMedian 1.3 s

Input / 1M—Median $0.60

Output / 1M—Median $2.55

Context128KMedian 262K

Elo—Median 1422

Capabilities

Out of 100

AgenticNone counted——

CodingNone counted——

Reasoning1 counted13.3—

MultimodalNone counted——

Knowledge2 counted31.9—

MultilingualNone counted——

Instruction1 counted10.4—

Math1 counted41.2—

Results

5 counted

BenchmarkResultPlace

GPQA diamond · Epoch AIKnowledge64.1% ±2.8125 / 217

OTIS Mock AIME 2024-2025Math55.6% ±6.2116 / 200

Artificial Analysis GPQA DiamondKnowledge61.5%155 / 199

Artificial Analysis Humanity's Last ExamKnowledge4.6%185 / 222

Artificial Analysis Long Context ReasoningReasoning8.7%199 / 217

Artificial Analysis IFBenchInstruction22.9%144 / 146

Sources

BenchLMCC BY-NC 4.0 Epoch AICC BY

Questions

How fast is DeepSeek R1 Distill Qwen 32B?: 60 output tokens per second, 0.8 s to the first token.
What is the context window of DeepSeek R1 Distill Qwen 32B?: 128K tokens.
Is DeepSeek R1 Distill Qwen 32B open source?: Yes. Its weights are published.
When was DeepSeek R1 Distill Qwen 32B released?: January 2025.

More from DeepSeek

All models

DeepSeek-R138.9 R1 Distill Llama 70BUnranked DeepSeek V329.9 DeepSeek V3 032437.1 DeepSeek-R1-052844.6