Home  /  Models  /  DeepSeek V3
⚡ Quick Pricing Summary

DeepSeek V3 by DeepSeek costs $0.140 per 1 Million input tokens and $0.280 per 1 Million output tokens. It features a 131k context window and has a blended 3:1 production rate of $0.175/1M tokens.

DeepSeek Flagship / Open-Weights

DeepSeek V3

671B MoE (37B active) open-weights powerhouse providing top-tier intelligence at disruptive pricing ($0.14/1M input).

Calculate Spend in App →
Input Token Price
$0.140
Per 1,000,000 tokens
Output Token Price
$0.280
Per 1,000,000 tokens
Prompt Caching Rate
$0.014
Save up to 80% on cached inputs
Max Context Window
131k
Max output: 8k tokens

Verified Specifications & Benchmark Data

Developer / Provider DeepSeek
Model Family DeepSeek
Blended 3:1 Rate (Production Benchmark) $0.175 / 1M tokens
Batch API Discount (24hr SLA) None
Multimodal Vision ❌ Text Only
Function Calling / Structured Outputs ✅ Native Tool Calling
MMLU Benchmark Score 88.5%
Latency & Throughput Tier Fast (~60 t/s)
Optimal Architecture & Use Cases General LLM routing, high-scale automation, cost reduction vs OpenAI

Frequently Asked Questions about DeepSeek V3

How much does DeepSeek V3 cost per 1M tokens?

DeepSeek V3 pricing is set at $0.140 per 1 million input tokens and $0.280 per 1 million output tokens. For high-volume batch workloads, 24-hour batch queues provide a None.

How much money does DeepSeek V3 prompt caching save?

With prompt caching enabled, cached input tokens are discounted to $0.014/1M, saving 90% on repeated system prompts and document vectors.

What is the context window limit of DeepSeek V3?

DeepSeek V3 has a maximum context window of 131k (131,072 tokens), supporting up to 8,192 completion tokens per response.

Direct Matchups with DeepSeek V3

See how DeepSeek V3 compares against other leading frontier and open-weights models in cost and latency.