DeepSeek V3.2 Exp API Pricing 2026

Overview

DeepSeek V3.2 Exp was released on September 29, 2025. Pricing starts at $0.210 per million input tokens and $0.320 per million output tokens. The model supports a context window of up to 164K tokens. API access is available through Deepseek.

Context Window

164K

tokens

Pricing

Input$0.210

Output$0.320

per 1M tokens

Speed

Output30 tok/s

TTFT1.28s

median latency

Capabilities

Tools

Pricing Comparison

Compare DeepSeek V3.2 Exp with 10 similar models by price.

Current Pricing (per 1M tokens)

11 models

Provider	Model	Input $/M	Output $/M	Context	Actions
DS Deepseek	deepseek-v3.2-exp	$0.210	$0.320	163,840	Try
21 Ai21	jamba-mini-1.7	$0.200	$0.400	256,000	Try
MM Minimax	minimax-01	$0.200	$1.100	1,000,192	Try
MM Minimax	minimax-m2	$0.200	$1.000	196,608	Try
QW Qwen	qwen3-vl-30b-a3b-thinking	$0.200	$1.000	131,072	Try
QW Qwen	qwen3-vl-235b-a22b-instruct	$0.200	$1.200	262,144	Try
QW Qwen	qwen-2.5-vl-7b-instruct	$0.200	$0.200	32,768	Try
MI Mistral AI	mistral-7b-instruct-v0.3	$0.200	$0.200	32,768	Try
M Meta-llama	llama-guard-2-8b	$0.200	$0.200	8,192	Try
MI Mistral AI	mistral-7b-instruct-v0.2	$0.200	$0.200	32,768	Try
QW Qwen	qwen-vl-plus	$0.210	$0.630	7,500	Try

* Some models use tiered pricing based on prompt length. Displayed prices are for prompts ≤ 200k tokens.

Compare Providers

DeepSeek V3.2 Exp is available from multiple providers with different pricing and availability.

5

Providers

$0.210

Cheapest Input/M

$0.320

Cheapest Output/M

AtlasCloud

Best Price

Provider Pricing

5 models

Provider	Input $/M	Output $/M	Uptime
A AtlasCloud Best Price	$0.210	$0.320	59.8%
D DeepInfra Best Price	$0.210	$0.320	99.9%
S SiliconFlow	$0.270	$0.410	97.8%
OpenRouter	$0.270	$0.400	-
N Novita	$0.270	$0.410	100.0%

Provider pricing data sourced from OpenRouter and Helicone

Cost vs. Quality

Compare DeepSeek V3.2 Exp's benchmark performance against all models.

X-axis:

Y-axis:

Loading chart...

DeepSeek V3.2 Exp

Other models

Detailed Benchmark Scores

Intelligence

28.1 (79th pct)

Coding

28.9 (80th pct)

Math

57.7 (67th pct)

MMLU Pro

83.6 (84th pct)

GPQA

73.8 (79th pct)

LiveCodeBench

55.4 (68th pct)

AIME

-

Aider

74.2 (83th pct)

Benchmark data from Artificial Analysis and HuggingFace Open LLM Leaderboard

Try DeepSeek V3.2 Exp

Use our Calculator

Estimate your costs based on expected token usage.

Open Cost Calculator

Try in Playground

Test DeepSeek V3.2 Exp directly in your browser.

Open Playground