DeepSeek logo

DeepSeek R1 0528 Qwen3 8B

8B

by DeepSeek

DeepSeek-R1-0528 is a lightly upgraded release of DeepSeek R1 that taps more compute and smarter post-training tricks, pushing its reasoning and inference to the brink of flagship models like O3 and Gemini 2.5 Pro. It now tops math, programming, and logic leaderboards, showcasing a step-change in depth-of-thought. The distilled variant, DeepSeek-R1-0528-Qwen3-8B, transfers this chain-of-thought into an 8 B-parameter form, beating standard Qwen3 8B by +10 pp and tying the 235 B “thinking” giant on AIME 2024.

Chat with DeepSeek R1 0528 Qwen3 8B

Pricing

Input Tokens
Per 1M tokens
Free
Output Tokens
Per 1M tokens
Free
Image Processing
Per 1M tokens
$0.00/1M tokens

Supported Modalities

Input

text

Output

text

Performance Benchmarks

Intelligence Index
Overall intelligence score
16.4
Coding Index
Programming capability
7.7
Math Index
Mathematical reasoning
63.7
GPQA
Graduate-level questions
61.2%
MMLU Pro
Multitask language understanding
73.9%
HLE
Human-like evaluation
5.6%
LiveCodeBench
Real-world coding tasks
51.3%
AIME 2025
Advanced mathematics
63.7%
MATH 500
Mathematical problem solving
93.2%

Specifications

Context Length
128K tokens
Provider
DeepSeek
Throughput
64.83 tokens/s
Released
May 29, 2025
Model ID
deepseek/deepseek-r1-0528-qwen3-8b

Ready to try it?

Start chatting with DeepSeek R1 0528 Qwen3 8B right now. No credit card required.

Start Chatting

More from DeepSeek

View all models