deepseek-reasoner

completions

bydeepseek

deepseek-reasoner

completions

Published by: deepseekProvider:

deepseek

DeepSeek-V3.1 (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

Aug 21, 2025

Knowledge

Feb 22, 2025

Context

128K

Input

$0.56 / 1M tokens

Output

$1.68 / 1M tokens

Cached

$0.07 / 1M tokens

Accepts: text

Returns: text

Released Aug 21, 2025Knowledge Cutoff: Feb 22, 2025

Context: 128K Input: $0.56 / 1M tokensOutput: $1.68 / 1M tokensCached: $0.07 / 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:

Academia: Score 65.95 (Top 8% - Rank #28)
Finance: Score 71.85 (Top 8% - Rank #28)
Marketing: Score 59.01 (Top 11% - Rank #38)
Maths: Score 89.70 (Top 7% - Rank #25)
Programming: Score 58.80 (Top 3% - Rank #11)
Science: Score 73.49 (Top 12% - Rank #42)
Writing: Score 57.96 (Top 19% - Rank #66)

Overall Performance: 68.10841269841269 average score across all categories

Detailed Benchmark Scores:

Benchmark	Score	Percentile	Domain
AA Coding Index	58.80	Top 3%	Programming
AAII	54.00	Top 6%	General
AA Math Index	89.70	Top 7%	Mathematics
GPQA	77.90	Top 14%	STEM (Physics, Chemistry, Biology)
HLE	13.00	Top 8%	General Knowledge
LiveCodeBench	78.40	Top 2%	Programming
MMLU-Pro	85.10	Top 3%	General Knowledge
SciCode	39.10	Top 8%	Scientific

GPQA Score: 77.90 - Graduate-level reasoning benchmark

Model Comparison:

Provider: deepseek

Model Type: completions

Context Size: 128000 tokens

Comparing against 345 models in the database

Performance Analytics for deepseek-reasoner

Usage Statistics (Last 6 Days):

Total Requests: 36 API calls
Average TPS: 192.43 tokens per second
Average Response Time: 47890.60ms
Average Time to First Token: 11371.60ms
Total Cost: $0.19
Average Request Cost: $0.0052

Daily Performance Breakdown:

Date	Requests	TPS	Response Time	TTFT	Cost
9/13/2025	1	0.00	860.00ms	860.00ms	$0.00
9/15/2025	4	0.00	732.30ms	732.30ms	$0.00
9/16/2025	1	0.00	1076.80ms	1076.80ms	$0.00
9/17/2025	4	0.00	839.40ms	839.40ms	$0.00
9/18/2025	16	205.87	80209.60ms	16021.20ms	$0.15
9/19/2025	10	156.21	43248.70ms	14481.60ms	$0.04

Performance Summary:

Model: deepseek-reasoner by deepseek

Monitoring Period: 9/13/2025 to 9/20/2025

Average Daily Requests: 6

Peak Daily Requests: 16

Statistics

View Full Statistics

Request Volume

Daily API requests

Performance (TPS)

Tokens per second

192.43 tokens/s

Category Scores

finance#10 maths#8 programming#5

Benchmark Tests

View Other Benchmarks

AA Coding Index

58.8

Programming

AAII

54.0

General

AA Math Index

89.7

Mathematics

GPQA

77.9

STEM (Physics, Chemistry, Biology)

HLE

13.0

General Knowledge

LiveCodeBench

78.4

Programming

MMLU-Pro

85.1

General Knowledge

SciCode

39.1

Scientific

Metric	AA Coding Index	AAII	AA Math Index	GPQA	HLE	LiveCodeBench	MMLU-Pro	SciCode
Score	58.8	54.0	89.7	77.9	13.0	78.4	85.1	39.1

Compare with Similar Models

gemini-2.5-pro-preview

grok-4

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