DeepSeekCertifiedReleased 12/01/2025

DeepSeek V3.2

DeepSeek‑V3.2 from DeepSeek harmonizes high computational efficiency with superior reasoning and agent performance. It builds on three main techniques: DeepSeek Sparse Attention for long‑context efficiency, a scalable reinforcement learning framework, and a large‑scale agentic task synthesis pipeline. This model excels at long-context reasoning and agentic tasks, efficiently handling extended inputs while maintaining strong accuracy. Its sparse attention design enables it to process complex, multi-step workflows without excessive compute costs. Overall, DeepSeek‑V3.2 targets long‑context reasoning, tool‑using agents, and efficient deployment in production environments.

View API reference ↓
Input / output priceFrom$0.26 / $0.38per 1M
24h uptime99.99%
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferencebureau.com/v1",
  apiKey: process.env.IB_API_KEY,
});

const res = await client.chat.completions.create({
  model: "deepseek/deepseek-v3.2",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

console.log(res.choices[0].message.content);

Providers

Dispatch routes requests across these providers and fails over between them without changing the model. Restrict or order providers with team Directives. Prices are gateway list prices before the platform fee.

ProviderContextMax outputLatencyThroughputInputOutputCacheCapabilitiesZDRNo trainingRegionsStatus
Amazon Bedrock128K8K0.8 s49 tps$0.62$1.85Read —Write —2 featuresYesYesGlobalCertified
DeepInfra164K8K0.4 s—$0.26$0.38Read $0.13Write —3 featuresNoYesGlobalCertified

Uptime

24 hours

Request success rate per provider.

Amazon Bedrock100.00%
DeepInfra99.99%

Throughput

last hour

P50 output tokens per second.

Amazon Bedrock49 tps
DeepInfra—

Latency

last hour

P50 time to first token.

DeepInfra0.4 s
Amazon Bedrock0.8 s

Getting started

Create a key on the API Keys page and set it as IB_API_KEY. See the quickstart.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferencebureau.com/v1",
  apiKey: process.env.IB_API_KEY,
});

const res = await client.chat.completions.create({
  model: "deepseek/deepseek-v3.2",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

console.log(res.choices[0].message.content);

Top-level parameters

Standard Chat Completions parameters work as documented. These have model-specific behavior.

ParameterTypeRequiredDescription
modelstringYesModel ID in the form creator/model, e.g. deepseek/deepseek-v3.2. Must be allowed by your team's Directives.
max_tokensnumberNoHard cap on generated tokens, and the basis for the balance reserved per request. DeepSeek V3.2 supports up to 8,000 output tokens.
reasoning{ effort: 'low' | 'medium' | 'high' }NoReasoning effort, mapped to the provider's native configuration. Reasoning tokens count toward output usage.
toolsTool[]NoFunction definitions the model can call. Tool calls are returned in choices[].message.tool_calls.
Supportedmax_tokenstemperaturestoptoolstool_choicereasoninginclude_reasoning

Input limits

InputSourcesLimits
Text—Prompt and response share the 128K-token context window

More models by DeepSeek

ModelContextInputOutputCacheRelease
deepseek/deepseek-v4.1-flash-1017-ultrafast1.05M$2.40$9.60Read $0.048Write —11/06/2026
deepseek/deepseek-v4.1-flash-10171.05M$0.30$1.20Read $0.0070Write —10/17/2026
deepseek/deepseek-v4.1-flash-1017-fast1.05M$0.60$2.40Read $0.014Write —10/17/2026
deepseek/deepseek-v4.1-flash-fast1.05M$0.60$2.40Read $0.014Write —09/29/2026
deepseek/deepseek-v4.1-flash1.05M$0.30$1.20Read $0.0060Write —09/08/2026
deepseek/deepseek-v4-flash-vision-exp1.05M$0.22$0.65Read $0.0068Write —08/21/2026
deepseek/deepseek-v4-pro-08131M$0.66$1.98Read $0.066Write —08/12/2026
deepseek/deepseek-v4-flash1M$0.13$0.26Read $0.028Write —04/23/2026
deepseek/deepseek-v4-flash-07311M$0.13$0.26Read $0.028Write —04/23/2026
deepseek/deepseek-v4-pro1.05M$1.32$3.96Read $0.13Write —04/23/2026