Hormouz AIHormouzSign in

DeepSeek: R1 0528: AI model

DeepSeek R1 0528 is the update to the open-source DeepSeek R1 reasoning model, improving reasoning and answer quality while keeping 671B parameters (37B active), via Hormouz.

Context window:
163,840 tokens
Input:
$0.5/1M tokens
Output:
$2.15/1M tokens
Type:
text
  • Tools
  • Streaming
  • Long context

Thinking levels: Always thinks

DeepSeek R1 0528 is the updated version of DeepSeek R1, with improved reasoning, accuracy and answer quality. Like the original, it is built with 671 billion parameters (37 billion active per pass) and open reasoning tokens, and it is available open-source.

This release advances especially on hard reasoning, math and coding tasks, and it fits problems that call for step-by-step thinking. Its context window is wide and works for long documents and long reasoning chains.

Pricing is usage-based: $0.70/M in · $2.50/M out. Hormouz provides stable access to the model from inside Iran, with no foreign credit card or circumvention tooling required.

May 28th update to the original DeepSeek R1 Performance on par with OpenAI o1, but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-r1-0528",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

VendorDeepSeek
TypeText
Context window163,840 tokens
Pricing$0.50/M in · $2.15/M out
APIChat Completions
CapabilitiesStreaming · Tools

Benchmarks

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
81.3%
HLE
Humanity's Last Exam
15.8%
IFBench
Instruction-following
39.6%
AA-LCR
Long-context reasoning
55.7%
CritPt
Research-level physics reasoning
1.4%
AA Omniscience
Broad factual knowledge accuracy
30.5%
Non-Hallucination
Rate of avoiding fabricated answers
16.6%

Coding

Terminal-Bench Hard
Agentic coding & terminal use
15.9%

Agentic

τ²-Bench
Tool-use agents in dual-control scenarios
36.5%

Source: Artificial Analysis

3D
Top 63%
1,122Elo
Win rate: 53.3%Avg time: 68.6s
Code
Top 66%
1,144Elo
Win rate: 52.6%Avg time: 88.6s
Data Viz
Top 47%
1,190Elo
Win rate: 60.9%Avg time: 76.0s
Game Dev
Top 69%
1,110Elo
Win rate: 49.3%Avg time: 107s
SVG
Top 84%
1,038Elo
Win rate: 48.7%Avg time: 60.2s
UI Component
Top 70%
1,109Elo
Win rate: 54.9%Avg time: 66.4s
Website
Top 62%
1,156Elo
Win rate: 52.7%Avg time: 89.8s

Metrics sourced from Design Arena

Code example

cURL
curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-r1-0528",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
Python
from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="deepseek/deepseek-r1-0528",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
JavaScript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "deepseek/deepseek-r1-0528",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the DeepSeek: R1 0528 API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches DeepSeek: R1 0528 and 600+ other models, and you pay only for what you use.
  • No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Create an accountBuy DeepSeek API access, paid in rial

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does this model cost?
Billing is usage-based: $0.70 per million input tokens and $2.50 per million output tokens. There is no fixed monthly fee. You pay only for what you use.
How is DeepSeek R1 0528 different from DeepSeek R1?
R1 0528 is the updated version of R1 with improved reasoning, accuracy and answer quality, while keeping the same open-source architecture with open reasoning tokens.
What is DeepSeek R1 0528 good for?
Complex reasoning problems, mathematics, coding and tasks that benefit from step-by-step thinking.
How do I access this model from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides a stable path to the model from inside Iran, with no foreign credit card required.
Is this model compatible with OpenAI code?
Yes. Hormouz exposes an OpenAI-compatible API. Just change the base URL to Hormouz and swap in your Hormouz key, and your existing SDKs run without a rewrite.

Updated 2026-10-13