ark-deepseek-v3-250324

Common Name: DeepSeek V3 (250324)

ByteDance
Released on Feb 17, 2025 12:00 AMSupportedTool Invocation
CompareTry in Chat

DeepSeek V3 model hosted on ByteDance platform with 64K context support. A powerful Mixture-of-Experts model delivering strong performance across coding and reasoning benchmarks.

Specifications

Context
64K
Maximum Output
16.4K
Inputtext
Outputtext

Performance (7-day Average)

Collecting…
Collecting…
Collecting…

Pricing

Input¥2.00/MTokens
Output¥8.00/MTokens

Availability Trend (24h)

Performance Metrics (24h)

Similar Models

¥4.00/¥16.00/M
ctx128Kmax16Kavailtps

DeepSeek's reasoning-focused model hosted on ByteDance infrastructure, optimized for complex problem-solving and logical reasoning tasks. Supports 128K context with strong analytical capabilities.

¥0.80/¥8.00/M
ctx128Kmaxavailtps
InOutCap

A vision-focused Doubao Seed 1.6 model for text, image, and video understanding.

¥0.80/¥2.00/M
ctx128Kmax16Kavailtps
InOutCap

A legacy Doubao 1.5 Pro model tuned for reliable text generation within a 128K context window.

¥0.15/¥1.50/M
ctx256Kmax16Kavailtps
InOutCap

A low-latency Doubao Seed 1.6 variant for cost-sensitive multimodal chat and tool workflows.