NOUMENOpen app
NOUMEN

DeepSeek · Open weights

DeepSeek V3.1 Thinking

Added Jan 1, 2024Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Thinking enabled version of Deepseek V3.1. DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode. It does better at tool calling and agent tasks, and has higher thinking efficiency than its predecessor. Quantized at FP8.

  • Tools
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.20
Output
$0.70

Reference prices as published by the model provider. API example

Specifications

Context window
128K
Max output
65.5K
Average output (7 days)
483 tokens
Longer than 48% of 468 models

Side by side

DeepSeek V3.1 Thinking compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

DeepSeek V3.1 ThinkingDeepSeekThis modelDeepSeek V3/Deepseek ChatDeepSeekRemove
ProviderDeepSeekDeepSeek
Discovery tierOpen weightsOpen weights
AddedJan 1, 2024Feb 27, 2025
Limits
Context window128K128K
Max output65.54KBest8.19K
Average output483 tokens293 tokens
Pricing, USD per 1M tokens
Input$0.20$0.10Best
Output$0.70$0.43Best
Performance
Throughput11.1 tok/s24.8 tok/sBest
Time to first token4.0 sBest4.3 s
Request success (24h)100.0%Best99.9%
Benchmark indexes
Intelligence Index13.516.0Best
Math Index89.7Best59.0
Capabilities
ToolsYesYes
Structured outputNoYes
PDF inputNoYes
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • DeepSeek V3.1 Thinking
  • DeepSeek V3/Deepseek Chat

Intelligence Index

  • DeepSeek V3.1 Thinking13.5
  • DeepSeek V3/Deepseek Chat16.0

DeepSeek V3.1 Thinking: Better than 34% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • DeepSeek V3.1 Thinking: 89.7Top
    • DeepSeek V3/Deepseek Chat: 59.0

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • DeepSeek V3.1 Thinking: 56.7%Top
    • DeepSeek V3/Deepseek Chat: 45.7%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • DeepSeek V3.1 Thinking: 14.3%Top
    • DeepSeek V3/Deepseek Chat: 11.2%

  • GPQA Diamond

    Graduate-level science questions

    • DeepSeek V3.1 Thinking: 77.9%Top
    • DeepSeek V3/Deepseek Chat: 75.1%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • DeepSeek V3.1 Thinking: 85.1%Top
    • DeepSeek V3/Deepseek Chat: 83.7%

  • IFBench

    Following precise output instructions

    • DeepSeek V3.1 Thinking: 41.5%
    • DeepSeek V3/Deepseek Chat: 49.0%Top

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • DeepSeek V3.1 Thinking: 78.4%Top
    • DeepSeek V3/Deepseek Chat: 59.3%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • DeepSeek V3.1 Thinking: 89.7%Top
    • DeepSeek V3/Deepseek Chat: 59.0%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • DeepSeek V3.1 Thinking: 25.0%
    • DeepSeek V3/Deepseek Chat: 32.6%Top

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • DeepSeek V3.1 Thinking: 37.4%
    • DeepSeek V3/Deepseek Chat: 78.9%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • DeepSeek V3.1 Thinking: 56.7%Top
    • DeepSeek V3/Deepseek Chat: 45.7%

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Throughput
11.1 tok/s
Faster than 1% of 254 models
Time to first token
4.0 s
Quicker than 44% of 254 models
Prompt cache hits
59.5%
Request success (24h)
100.0%
Reliable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 100.0%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/DeepSeek-V3.1:thinking",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V3.1:thinking",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

DeepSeek

Other DeepSeek models, starting with the same family.