DeepSeek

DeepSeek-V3

Generally availableOpen weights64K context

Summary

DeepSeek-V3 is an open-weight language model from DeepSeek, released on 26 Dec 2024. It accepts text and generates text, with a 64K-token context window and up to 8K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.32 per million input tokens and $0.89 per million output tokens (DeepInfra), across 3 providers we track.

DeepSeek chat model for instruction following, coding, and analysis

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperDeepSeek
Release date26 Dec 2024 [models.dev]
StatusGenerally available [models.dev]
WeightsOpen weights [models.dev]
Context window64K (64,000 tokens) [models.dev]
Max output8K tokens [models.dev]
Knowledge cutoffJul 2024 [models.dev]
InputText [models.dev]
OutputText [models.dev]
Reasoning modeNo [models.dev]
Tool callingYes [models.dev]
Structured outputYes [models.dev]

DeepSeek-V3 API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
DeepInfra
deepseek-ai/DeepSeek-V3
$0.32$0.89–164Kmodels.dev, 18 Nov 2026
Hugging Face Inference Providers
deepseek-ai/DeepSeek-V3
$0.40$1.30–64Kmodels.dev, 18 Nov 2026
Novita AI
deepseek/deepseek_v3
$0.89$0.89–64Kmodels.dev, 18 Nov 2026

Frequently asked questions

What is the context window of DeepSeek-V3?

DeepSeek-V3 has a context window of 64,000 tokens (64K) and can generate up to 8,192 tokens in a single response, according to models.dev.

How much does the DeepSeek-V3 API cost?

DeepSeek-V3 is listed from $0.32 per million input tokens and $0.89 per million output tokens (DeepInfra). Prices last verified 18 Nov 2026.

Is DeepSeek-V3 open source?

DeepSeek-V3 is an open-weight model. Check the license terms before commercial use.

When was DeepSeek-V3 released?

DeepSeek released DeepSeek-V3 on 26 Dec 2024, according to models.dev.

Which providers offer DeepSeek-V3?

We track DeepSeek-V3 on 3 providers: DeepInfra, Hugging Face Inference Providers, Novita AI.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Skyfall 36B V2Thedrummer10 Mar 202533K$0.55$0.80Open
Qwen2.5-VL 72B InstructAlibaba (Qwen)26 Jan 2025128K$0.80$1Open
Llama-4-Maverick-17B-128E-Instruct-FP8Meta5 Apr 2025128K$0.24$0.85Open
Qwen3-VL 30B-A3BAlibaba (Qwen)Apr 2025131K$0.20$0.80Open
WizardLM 2 8x22BMicrosoft15 Apr 202466K$0.62$0.62Open
Aya Expanse 32BCohere24 Oct 2024128K$0.50$1.50Open

Change history

  • Knowledge cutoff changed from Dec 2024 to Jul 2024
  • Max output changed from 16K to 8K
  • Knowledge cutoff changed from Jul 2024 to Dec 2024
  • Knowledge cutoff changed from Dec 2024 to Jul 2024
  • Context window changed from 128K to 64K
  • Name changed from DeepSeek V3 to DeepSeek-V3
  • Max output changed from 8K to 16K
  • Open weights changed from no to yes
  • Name changed from DeepSeek-V3 to DeepSeek V3
  • Open weights changed from yes to no
  • Context window changed from 164K to 128K
  • Name changed from DeepSeek V3 to DeepSeek-V3
  • Knowledge cutoff changed from Jul 2024 to Dec 2024
  • Context window changed from 131K to 164K
  • Knowledge cutoff changed from Dec 2024 to Jul 2024
  • Open weights changed from no to yes
  • Context window changed from 128K to 131K
  • Context window changed from 131K to 128K
  • Open weights changed from yes to no
  • Knowledge cutoff changed from Jul 2024 to Dec 2024
  • Name changed from DeepSeek-V3 to DeepSeek V3
  • Context window changed from 164K to 131K
  • Context window changed from 64K to 164K
  • Context window changed from 128K to 64K
  • Knowledge cutoff changed from Dec 2024 to Jul 2024

Sources