DeepSeek › DeepSeek V3.2 › Pricing

DeepSeek V3.2 pricing

API list price by DeepSeek · checked Oct 26, 2026
Benchmarks and alternatives →
$0.28 / $0.40per 1M tokens in / out
$0.028cache read · −90%
#58on OpenRouter by tokens · 277.9B / week

Price sheetper 1M tokens

Input$0.28list
Output$0.40
Cache read$0.028−90% vs input
Context · max output128K · 164Ktokens

Where to buy12 providers · per 1M tokens

ProviderInputOutputContext
DeepInfra$0.21$0.32—
SiliconFlow$0.26$0.80164K
AtlasCloud$0.26$0.38164K
Featherless$0.26$0.41—
Baidu$0.28$0.42131K
deepseeklab$0.28$0.40—
DigitalOcean$0.30$0.96164K
Friendli$0.50$1.50164K
AWS Bedrock$0.62$1.85—
Phala$1$1164K
Mara$3$4.5033K
SambaNova$3$4.5033K

In DeepSeek's line-upper 1M in / out

DeepSeek OCR 2$0.03 / $0.03−90%
DeepSeek OCR$0.03 / $0.10−85%
DeepSeek R1 Distill Llama 8B$0.20 / $0.32−26%
DeepSeek R1 Distill Qwen 32B$0.27 / $0.27−13%
DeepSeek V4.1 Flash 1017$0.30 / $1.20+69%
DeepSeek R1 Distill Qwen 14B$0.43 / $1.12+95%
R1 0528$0.39 / $1.632.2×
<

Cheaper for the same workvs DeepSeek V3.2

Claude Haiku 5.5 max-35%score ×2.7$0.21
MiniMax M3.1 Flash thinking-15%score ×2.7$0.22
GLM 5.3 Flash maxOpen-23%score ×2.6$0.28
Qwen3.8 Flash thinkingOpen-26%score ×2.5$0.37
GPT-6 Luna max-35%score ×2.4$0.068

Pricing questions

How much does DeepSeek V3.2 cost?›

$0.28 per 1M input tokens and $0.40 per 1M output tokens. Cached input is $0.028.

Is DeepSeek V3.2 free to use?›

There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.

How much does prompt caching save?›

Cache reads cost $0.028 per 1M, 90% below the input price.

Is DeepSeek V3.2 cheaper than DeepSeek V3.1 Terminus?›

Yes: $0.28 / $0.40 against $0.27 / $0.95, 30% less per token.

What is a cheaper alternative to DeepSeek V3.2?›

Claude Haiku 5.5 at max effort costs -35% per task with a score ×2.7 on .