DeepSeek

platform.deepseek.com
+
+
+
+
Platform Stats
Total Models5
Organizations1
Verified Benchmarks0
Multimodal Models0
+
+
+
+
Pricing Overview
Avg Input (per 1M)$0.38
Avg Output (per 1M)$1.54
Cheapest Model$0.27
Premium Model$0.55
+
+
+
+
Supported Features
Number of models supporting each feature
web Search
0
function Calling
3
structured Output
0
code Execution
0
batch Inference
0
finetuning
0
+
+
+
+
Input Modalities
Models supporting different input types
text
5 (100%)
image
0 (0%)
audio
0 (0%)
video
0 (0%)
+
+
+
+
Models Overview
Top performers and pricing distribution

Pricing Distribution

Input pricing per 1M tokens
$0-1
5 models

Top Performing Models

By benchmark avg
#1DeepSeek-V3.2 Thinking
73.1%
#2DeepSeek-V3.2
59.0%
#3DeepSeek-R1
3.1%
#4DeepSeek-V3.1
1.3%

Most Affordable Models

DeepSeek-V3.2
$0.27/1M
DeepSeek-V3.2-Speciale
$0.27/1M
DeepSeek-V3.1
$0.27/1M

Available Models

5 models available through DeepSeek

LicenseLinks
#01DeepSeekDeepSeek-V3.2 Thinking
DeepSeek-V3.2-Exp, the experimental thinking release of DeepSeek-V3.2 with 685B parameters
DeepSeek
Sep 29, 2025
MIT
73.1%
#02DeepSeekDeepSeek-V3.2
DeepSeek-V3.2 with 685B parameters, agentic tool-use, thinking modes, and 128K context
DeepSeek
Dec 1, 2025
MIT
-
#03DeepSeekDeepSeek-V3.2-Speciale
DeepSeek-V3.2-Speciale, a high-compute variant of V3.2 with gold-medal reasoning performance
DeepSeek
Dec 1, 2025
MIT
-
#04DeepSeekDeepSeek-V3.1
DeepSeek-V3.1 with 671B parameters, hybrid thinking and non-thinking modes, and 128K context
DeepSeek
Aug 21, 2025
MIT
-
#05DeepSeekDeepSeek-R1
DeepSeek's R1 reasoning model with 671B parameters, 128K context, and extended chain-of-thought
DeepSeek
Jan 20, 2025
MIT
-
+
+
+
+
Resources
Official Website