All Comparisons

DeepSeek-V3 vs o3

DeepSeek-V3
DeepSeek 🇨🇳
23.4
#265
low
o3
OpenAI 🇺🇸
47.9
#174
high
Best Benchmark Score
o3
47.9/100

Model Specifications

SpecDeepSeek-V3o3
ProviderDeepSeekOpenAI
Release DateDec 2024Apr 2025
Knowledge Cutoff2024-05-31
Parameters671000000000BUndisclosed
Context Window
Max Output
Open SourceYesNo
Licensemit_+_model_license_(commercial_use_allowed)proprietary
Tokenizer
Modality
Reasoning ModelNoNo
ModeratedNoNo

Pricing Comparison

Price (per 1M tokens)DeepSeek-V3o3

API Provider Pricing

Capabilities

FeatureDeepSeek-V3o3
Text Input
Image Input (Vision)
Audio Input
Video Input
File Input
Image Output
Audio Output
Tool Use
Structured Output (JSON)
Streaming
Reasoning Tokens
Web Search
Temperature Control
Top-P Sampling
Stop Sequences
Seed (Reproducibility)
Log Probabilities

Category Comparison

Benchmark-by-Benchmark

general

BenchmarkDeepSeek-V3o3
SimpleQA
24.9

coding

BenchmarkDeepSeek-V3o3
SWE-bench Verified
42
69.1

reasoning

BenchmarkDeepSeek-V3o3
GPQA Diamond
59.1
83.3
Humanity's Last Exam
14.7
ARC-AGI v2
6.5

math

BenchmarkDeepSeek-V3o3
AIME 2025
86.4
FrontierMath
15.8

multimodal

BenchmarkDeepSeek-V3o3
MMMU
82.9
MMMU-Pro
76.4
CharXiv Reasoning
78.6

agent

BenchmarkDeepSeek-V3o3
BrowseComp
49.7

Category Winners

coding
o3
58.4
math
o3
43.5
reasoning
o3
34.1
general
DeepSeek-V3
30.5
multimodal
o3
73.4
agent
o3
23.4

Compare More Models

Frequently Asked Questions