deepseek-reasoner

completions

bydeepseek

deepseek-reasoner

completions

Published by: deepseekProvider:

deepseek

DeepSeek-V3.2-Exp (Thinking Mode) is an advanced AI model designed to enhance logical reasoning and problem-solving capabilities, leveraging deep learning techniques to provide accurate and contextually relevant insights across various domains.

Released

Sep 29, 2025

Knowledge

Feb 22, 2025

License

MIT

Context

128K

Input

$0.28 / 1M tokens

Output

$0.42 / 1M tokens

Cached

$0.03 / 1M tokens

Accepts: text

Returns: text

Released Sep 29, 2025Knowledge Cutoff: Feb 22, 2025License: MIT

Context: 128K Input: $0.28 / 1M tokensOutput: $0.42 / 1M tokensCached: $0.03 / 1M tokensAccepts: textReturns: text

Access deepseek-reasoner through LangDB AI Gateway

Recommended

Integrate with deepseek's deepseek-reasoner and 250+ other models through a unified API. Monitor usage, control costs, and enhance security.

Unified API

Cost Optimization

Enterprise Security

Get Started Now

Free tier available • No credit card required

Instant Setup

99.9% Uptime

10,000+Monthly Requests

Benchmark Results for deepseek-reasoner

Category Performance Scores:

Maths: Score 92.00 (Top 6% - Rank #20)
Finance: Score 78.95 (Top 6% - Rank #20)
Science: Score 77.76 (Top 7% - Rank #23)
Writing: Score 67.64 (Top 10% - Rank #33)
Academia: Score 74.95 (Top 4% - Rank #13)
Marketing: Score 69.90 (Top 7% - Rank #23)
Programming: Score 52.80 (Top 4% - Rank #13)

Overall Performance: 73.42865079365079 average score across all categories

Detailed Benchmark Scores:

Benchmark	Score	Percentile	Domain
HLE	22.20	Top 5%	General Knowledge
GPQA	84.00	Top 4%	STEM (Physics, Chemistry, Biology)
SciCode	38.90	Top 20%	Scientific
MMLU-Pro	85.60	Top 7%	General Knowledge
LiveCodeBench	86.20	Top 4%	Programming
AA Math Index	92.00	Top 6%	Mathematics
AA Coding Index	52.80	Top 4%	Programming
AAII	65.90	Top 4%	General

GPQA Score: 84.00 - Graduate-level reasoning benchmark

Model Comparison:

Provider: deepseek

Model Type: completions

Context Size: 128000 tokens

Comparing against 322 models in the database

Performance Analytics for deepseek-reasoner

Usage Statistics (Last 3 Days):

Total Requests: 63 API calls
Average TPS: 465.79 tokens per second
Average Response Time: 44690.20ms
Average Time to First Token: 10130.90ms
Total Cost: $0.19
Average Request Cost: $0.0031

Daily Performance Breakdown:

Date	Requests	TPS	Response Time	TTFT	Cost
11/11/2025	7	687.22	20708.30ms	4793.00ms	$0.01
11/13/2025	21	403.72	67341.80ms	26011.60ms	$0.12
11/14/2025	35	510.12	35895.50ms	1670.10ms	$0.06

Performance Summary:

Model: deepseek-reasoner by deepseek

Monitoring Period: 11/11/2025 to 11/18/2025

Average Daily Requests: 21

Peak Daily Requests: 35

Statistics

View Full Statistics

Request Volume

Daily API requests

Performance (TPS)

Tokens per second

465.79 tokens/s

Category Scores

maths#9 finance#8 academia#8 programming#6

Benchmark Tests

View Other Benchmarks

HLE

22.2

General Knowledge

GPQA

84.0

STEM (Physics, Chemistry, Biology)

SciCode

38.9

Scientific

MMLU-Pro

85.6

General Knowledge

LiveCodeBench

86.2

Programming

AA Math Index

92.0

Mathematics

AA Coding Index

52.8

Programming

AAII

65.9

General

Metric	HLE	GPQA	SciCode	MMLU-Pro	LiveCodeBench	AA Math Index	AA Coding Index	AAII
Score	22.2	84.0	38.9	85.6	86.2	92.0	52.8	65.9

Compare with Similar Models

claude-opus-4.5

gemini-3-flash-preview

Code Examples

Integration samples and API usage

Code Samples for deepseek-reasoner

Python SDK Example:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.us-east-1.langdb.ai//projects/<your_project_id>",
    api_key="<your_api_key>"
)

response = client.chat.completions.create(
    model="deepseek-reasoner",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

TypeScript SDK Example:

import OpenAI from 'openai';

const client = new OpenAI({
    baseURL: "https://api.us-east-1.langdb.ai//projects/<your_project_id>",
    apiKey: "<your_api_key>"
});

const response = await client.chat.completions.create({
    model: "deepseek-reasoner",
    messages: [
        { role: "user", content: "Hello, how are you?" }
    ]
});

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

cURL Example:

curl -X POST "https://api.us-east-1.langdb.ai//projects/<your_project_id>/v1/chat/completions" \
  -H "Authorization: Bearer <your_api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-reasoner",
    "messages": [
        {"role": "user", "content": "Hello, how are you?"}
    ]
  }'

Model: deepseek-reasoner

Provider: deepseek

API Endpoint: $https://api.us-east-1.langdb.ai/

Create API Key

Related Models

Similar models from deepseek

deepseek-reasoner by deepseek - AI Model Details, Pricing, and Performance Metrics

deepseek-reasoner

deepseek-reasoner

Access deepseek-reasoner through LangDB AI Gateway

Statistics

Category Scores

Benchmark Tests

Compare with Similar Models

Code Examples

Related Models

deepseek-chat