NVIDIA · current

DeepSeek V3.2

DeepSeek V3.2 is attributed to NVIDIA in the catalog. It accepts text input and produces text output. Reported capabilities include reasoning and tool calling.

Compare model

Report an error Model and price history

Modalities

Token rates

Unavailable

No recent paired token rate available

Context

131K

Released

2025-12-01

Knowledge cutoff

Unknown

Where to use it

1 observed offerings · token prices in USD per million where available

Estimate your workload cost

Vultr

nvidia/DeepSeek-V3.2-NVFP4

current

Input

$0.55

Output

$1.65

Cache

—

Context

131K

Max output

131K

Reasoning: Yes
Tool calling: Yes
Structured output: No
Temperature: Unknown

Historical price

Limits and capabilities reflect each provider’s listed offering. Unknown means support was not reported.

Provider / API ID	Status	Context	Max output	Input	Output	Cache read	Reasoning	Tools	Structured	Temperature
Vultr nvidia/DeepSeek-V3.2-NVFP4	currentHistorical price	131K	131K	$0.55	$1.65	—	Yes	Yes	No	Unknown

Capabilities and limits

Input: text
Output: text
Max output: 131K
Weights: Open (source-reported)
Release: 2025-12-01
Knowledge cutoff: Unknown
Family: deepseek

Reasoning · Tool calling

Provider-specific limits and support can differ.

Plan a model evaluation

Evaluate quality and latency on examples from your intended use before choosing a model.

Explore available evidence

Models to compare

Candidates match reported specifications; they are not tested replacements.

V3.2 ExpDeepSeek · 164K contextCompare →ai/DeepSeek-V3.2DeepSeek · 164K contextCompare →V4.1 Flash 1017 UncensoredDeepSeek · 1.05M contextCompare →

Model and price history

Price history (6 observations)

Date	Provider	Input / output
2026-09-22	Vultr	$0.55 / $1.65
2026-09-21	Vultr	$0.55 / $1.65
2026-09-20	Vultr	$0.55 / $1.65
2026-09-19	Vultr	$0.55 / $1.65
2026-09-18	Vultr	$0.55 / $1.65
2026-09-17	Vultr	$0.55 / $1.65