deepseek Open weights

DeepSeek V3 0324 pricing & benchmarks

Released Mar 24, 2025 · served by 4 providers

Input / 1M tokens$0.240
Output / 1M tokens$0.900
Context window164K
Cached input$0.135
Intelligence index15.4
Coding index21.2
Agentic index1.5
1M in + 300K out$0.51

What DeepSeek V3 0324 is

DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team. It succeeds the [DeepSeek V3](/deepseek/deepseek-chat-v3) model and performs really well...

Providers and prices for DeepSeek V3 0324

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
DeepInfracheapest $0.240 $0.900 164K fp4 80.4%
SiliconFlow $0.250 $1.00 164K fp8 91.5%
Novita $0.270 $1.12 164K fp8 97.4%
Crusoe $0.500 $1.50 164K bf16 99.9%

Cheaper models in the same class

Models scoring within 4 points of DeepSeek V3 0324 on the Intelligence Index, but with a lower output price.

FAQ

How much does the DeepSeek V3 0324 API cost?

$0.240 per 1M input tokens and $0.900 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $0.51.

Is DeepSeek V3 0324 free?

No. It is a paid model starting at $0.240 per 1M input tokens, though some providers offer trial credits.

Which provider is cheapest for DeepSeek V3 0324?

DeepInfra at $0.240 input / $0.900 output per 1M tokens (fp4 quantization).

Can I self-host DeepSeek V3 0324?

Yes — weights are published as deepseek-ai/DeepSeek-V3-0324 on Hugging Face, so you can run it on your own hardware.