DeepSeek V3 (Dec '24)
DeepSeekDeepSeekOpen WeightMIT + Model License (Commercial use allowed)
Description
A powerful Mixture-of-Experts (MoE) language model with 671B total parameters (37B activated per token). Features Multi-head Latent Attention (MLA), auxiliary-loss-free load balancing, and multi-token prediction training. Pre-trained on 14.8T tokens with strong performance in reasoning, math, and code tasks.
Release Date
2024-12-26
Parameters
671.0B
Context Length
—
Modalities
text
Capability Radar
29
general
36
coding
38
reasoning
38
scienceest.
37
agents
0
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 304 | 29.0 | AA |
| General Ranking | 320 | 34.0 | AA |
| Math Reasoning | 226 | 39.0 | AA |
| Science | 297 | 39.0 | AA |
Benchmark Scores (LLM Stats)
Biology
GPQA
59.1%SR
Code
Aider-Polyglot Edit
79.7%SR
Aider-Polyglot
49.6%SR
SWE-Bench Verified
42.0%SR
LiveCodeBench
37.6%SR
Factuality
SimpleQA
24.9%SR
Finance
MMLU
88.5%SR
MMLU-Pro
75.9%SR
General
MMLU-Redux
89.1%SR
C-Eval
86.5%SR
IFEval
86.1%SR
CSimpleQA
64.8%SR
LongBench v2
48.7%SR
Language
CLUEWSC
90.9%SR
Math
DROP
91.6%SR
MATH-500
90.2%SR
CNMO 2024
43.2%SR
AIME 2024
39.2%SR
Reasoning
HumanEval-Mul
82.6%SR
FRAMES
73.3%SR
AA Evaluation Indices
Math Index26.0
Intelligence Index10.4
Math 5000.9
Mmlu Pro0.8
Gpqa0.6
Livecodebench0.4
Scicode0.4
Ifbench0.3
Lcr0.3
Aime 250.3
Aime0.3
Tau20.2
Terminalbench Hard0.1
Hle0.0
LLM Stats Category Scores
Instruction Following90
Language80
Legal80
Finance80
Healthcare80
Math70
Reasoning70
Structured Output70
General70
Physics60
Biology60
Chemistry60
Long Context50
Code50
Frontend Development40
Factuality20
Pricing
Input Price$0.4 / 1M tokens
Output Price$0.89 / 1M tokens
Blended Price (3:1)$0.523 / 1M tokens
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
17 providers
Cheapest: SiliconFlow (China)Most Expensive: Together AI
ProviderInputOutput
1SiliconFlow (China)Cheapest
$0.25
$1
2SiliconFlow
$0.25
$1
3Kilo Gateway
$0.27
$1
4Vercel AI Gateway
$0.27
$1.12
5NanoGPT
$0.28
$0.42
6Chutes
$0.28
$0.42
7D.Run (China)
$0.28
$1.1
8Alibaba (China)
$0.287
$1.147
9DeepSeekPRIMARY
$0.4
$0.89
10NovitaAI
$0.4
$1.3
11Nebius Token Factory
$0.4
$2
12Vultr
$0.55
$1.65
13Helicone
$0.56
$1.68
14Amazon Bedrock
$0.58
$1.68
15Vertex
$0.6
$1.7
16Synthetic
$1.25
$1.25
17Together AI
$1.25
$1.25
Compare pricing across different API providers for this model.