When was DeepSeek R1 (0528) released?

DeepSeek released DeepSeek R1 (0528) on 28 May 2025. It is available through the Opper gateway.

Does DeepSeek R1 (0528) support zero data retention?

Yes. DeepSeek R1 (0528) is available with zero data retention on DeepInfra, Novita through the Opper gateway.

Where is DeepSeek R1 (0528) hosted?

DeepSeek R1 (0528) runs on US hosting routes through the Opper gateway.

Is DeepSeek R1 (0528) trained on prompts and outputs?

No. Every hosting route for DeepSeek R1 (0528) is configured so customer prompts and outputs are not used to train models by default.

Is a GDPR DPA available for DeepSeek R1 (0528)?

Yes. A GDPR Data Processing Agreement is available on at least one route serving DeepSeek R1 (0528). Contact hello@opper.ai to receive it.

DeepSeek R1 (0528)

by DeepSeek

Versions:DeepSeek R1 (0528)DeepSeek R1

Announcement · All DeepSeek releases

DeepSeek R1 0528 is a major update to DeepSeek R1 released on May 28, 2025, adding system prompt support, JSON schema output, and native function calling for improved agentic compatibility. The model roughly doubles its reasoning token usage, averaging about 23K tokens per challenging problem versus 12K in the original, producing deeper chain-of-thought analysis and cutting hallucinations by 45 to 50% on rewriting, summarization, and reading comprehension tasks. Built on a 685-billion parameter MoE architecture with 37 billion active parameters, R1 0528 advances reasoning quality and approaches the performance of leading frontier reasoning models. It supports up to 163,840 tokens of context and outputs explicit thinking tokens for transparency. The model supports tool use and structured output via JSON schema, enabling multi-turn agentic workflows with improved reliability on complex tasks. Released under the MIT license, it remains fully open-source.

Call DeepSeek R1 (0528) on Opper with the OpenAI SDK. Sign up without a credit card. Get started

Get your API key Try in chat Compare

Key info

Input

Output

Features

Context window

164K

Max output

164K

Input price

$0.50 /1M$0.35 cache read

Output price

$2.15 /1M

Released

28 May 2025

Web search

Via Opper

US residency available
Zero data retention on pay-as-you-go
No training by default
GDPR DPA available

Available routes

DeepSeek R1 (0528) runs on 2 different routes through the Opper gateway. Compare residency, zero data retention and training posture at a glance, with full data-handling detail per route below. One route also sells flex processing: the same model at another price, picked per request with service_tier. How service tiers work.

Provider	Region	Zero data retention	Training	Input	Output	Cache read
DeepInfra	US		No	$0.50	$2.15	$0.35
Novita	US		No	$0.70	$2.50	$0.35	Not offered on this route

Live routes and prices as JSON: api.opper.ai/v3/models?q=deepseek-r1-0528

Uptime and availability

DeepSeek R1 (0528) runs on 2 independently monitored routes through the Opper gateway. A direct integration leaves you on one of them, the gateway serves whichever is healthy.

100%effective uptime, last 30 days

Averaged across these routes on their own, a single provider reached 99.93% over the same window.

Daily status through the gateway, which serves each request from whichever route is healthy.

2 monitored providers serve DeepSeek R1 (0528).

Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. See uptime for every provider Opper monitors.

Data handling per route

Each route hosting DeepSeek R1 (0528) has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.

DeepInfra — United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: No classifier
GDPR DPA: DPA available
Transfer mechanism: SCCs

Novita — United States🇺🇸

Zero data retention: nothing is logged, held for abuse monitoring or used for training. No training on customer data. US; SCCs; DPA available.

Zero data retention: Yes. Nothing is logged, held for abuse monitoring or used for training.
Training: No training on customer data.
Logging: None
Abuse monitoring: On by default, keeps nothing
GDPR DPA: DPA available
Transfer mechanism: SCCs

Best for

Where DeepSeek R1 (0528) ranks among the models on Opper with a score for each use case.

Multilingual#17 of 3586.0

Tasks across languagesSee the Multilingual ranking →

Benchmarks

Independent benchmark scores: the Intelligence Index and the individual evaluations behind it, grouped by skill.

Global rank#215 of 508 LLMs

TierEfficient

Output speed33 tok/s

Intelligence Index (previous version)13.1

Coding

LiveCodeBench

77%

Agentic & knowledge work

Terminal-Bench Hard

16%

τ²-Bench Telecom

37%

Science & reasoning

Humanity's Last Exam

16%

GPQA Diamond

81%

Long-context reasoning

56%

Knowledge

MMLU-Pro

85%

Math & instruction following

AIME 2025

76%

IFBench

40%

See full leaderboard →Benchmarks via Artificial Analysis · View on AA

Get started

Call DeepSeek R1 (0528) through the Opper gateway with one API key. Opper is drop-in compatible with the OpenAI, Anthropic and Google AI SDKs.

DeepSeek R1 (0528) is a premium model on Opper. Sign up needs no credit card: you get an API key straight away and the free models work in the playground and the API. Add a card to use premium models, pay-as-you-go with no minimum.

Set it up with your agent

Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up Opper for you.

Or call it directly

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.OPPER_API_KEY,
5  baseURL: "https://api.opper.ai/v3/compat",
6});
7
8const completion = await client.chat.completions.create({
9  model: "deepseek-r1-0528",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);