DeepSeekactiveOpen Source

DeepSeek-V3

deepseek-v3-0324

Highly capable open-source model rivaling proprietary alternatives at fraction of cost.

Context Window

131.1K

tokens

Max Output

8.2K

tokens

Input Price

$0.27

per 1M tokens

Output Price

$1.10

per 1M tokens

Details

Familydeepseek-v3

Parameters671B MoE

Training Cutoff2025-02-01

ReleasedMarch 24, 2025

Aliasesdeepseek-v3-0324

Capabilities

FunctionsStreamingJSON ModeCodeTool Use

Documentation

https://api.deepseek.com/chat/completions

Evaluation Scores(6 benchmarks)

HumanEvalFunction-level Python code generation

82.6%

MATH-500Competition-style math

80.4%

MMLU-ProHarder successor to MMLU

75.9%

GPQA DiamondPhD-level science questions

49.2%

SWE-bench VerifiedResolving real GitHub issues

38.8%

Quick Access

curl pikaainews.com/api/models/deepseek-v3

npx pika-models info deepseek-v3

Get API Access

Official

DeepSeek API

Official DeepSeek API. Cheapest frontier model access.

Third-Party Providers & Aggregators

Cerebras

Wafer-scale inference. 1000+ tokens/sec for select models.

DeepInfra

Lowest per-token rates for open-source models.

Fireworks AI

Fastest inference engine. Multimodal support, HIPAA/SOC2.

Groq

Ultra-fast LPU inference. Best latency for real-time apps.

OpenRouter

500+ models, one API key. Pay-per-token, no minimums.

SiliconFlow

China-optimized inference. Strong Qwen/DeepSeek support.

Together AI

Fast open-source model inference. Sub-100ms latency.

Other deepseek-v3 models

DeepSeek

DeepSeek-V3.2

deepseek-v3.2

131.1K ctx$0.27/1M

DeepSeek

DeepSeek-V3.1

deepseek-v3.1

131.1K ctx$0.27/1M