DeepSeek-V3

deepseek-v3

An open Mixture-of-Experts chat model for coding, math, and general reasoning.

Open DeepSeek MoE chat model for coding, math, and general reasoning

131K context131K context

Released2024-12-26

Relays5 sites

text generationcode generationtool useopen source

Model information

Model ID: deepseek-v3
Context: 131.1K
Max input: -
Max output: 8.2K
Knowledge cutoff: -
Open weights: Supported
License: DeepSeek Model License
Release date: 2024-12-26

Input

Text: SupportedImage: Not supportedAudio: Not supportedVideo: Not supportedPDF: Not supportedFile: Not supported

Output

Text: SupportedImage: Not supportedAudio: Not supportedVideo: Not supported

API capabilities

Reasoning: Not supportedTool call: SupportedStructured output: UnknownTemperature: Supported

Relay Comparison

Compare token, per-request, or per-second pricing by relay channel.

Unit: token prices are ¥ / 1M tokens; per-request and per-second rows use their shown unit

ClaudeCN

claudecn.topToken billing

Price range ¥1.8 / 1M tokens

Cache hit¥0.225

Channel	Billing	Input	Output	Cache hit	Cache write
国产模型	Token	¥1.8	¥7.2	¥0.225	-

RunAPI

https://runapi.hostToken billing

Price range ¥1.54 / 1M tokens

Cache hit¥0.3086

Channel	Billing	Input	Output	Cache hit	Cache write
default	Token	¥1.54	¥6.17	¥0.3086	-

AIGC Token

aigc-token.comToken billing

Price range ¥2 - ¥8 / 1M tokens

Channel	Billing	Input	Output	Cache hit	Cache write
AzurePlus	Token	¥2	¥8	¥0.4	-
default	Token	¥2	¥8	¥0.4	-
AzurePro	Token	¥8	¥32	¥1.6	-

猫云API

api.nekocloud.vipToken billing

Price range ¥2.5 / 1M tokens

Channel	Billing	Input	Output	Cache hit	Cache write
deepinfra	Token	¥2.5	¥7	¥2.5	¥0

ePhone AI

ephone.aiToken billing

Price range ¥1.6 / 1M tokens

Channel	Billing	Input	Output	Cache hit	Cache write
deepseek	Token	¥1.6	¥1.6	-	-

How should DeepSeek-V3 relay pricing be compared?

This DeepSeek-V3 pricing page compares official pricing with public prices from 5 listed AI gateways. Token prices are shown in CNY per 1M tokens, while per-request, per-second, and per-character rows use the unit shown in the table. Last updated: 10/23/2026, 08:14.

Data sources: Public price catalogs, official pricing records, and monitoring results.
Metric definitions: Uptime means successful probe response rate, fake-rate signals possible model mismatch or abnormal output risk, and latency is average API response time.
Risk note: Relay gateways are third-party services. Pricing, billing, privacy, and stability can change; start with a small top-up and verify reliability before continued use.