NOUMENOpen app
NOUMEN

DeepSeek · Open weights

DeepSeek V3.2

Added Dec 1, 2025Open weights

Compare models

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

Loading models…

Use in the app

DeepSeek V3.2 (non-thinking mode) — official successor to V3.2-Exp. Reasoning-first model built for agents with GPT-5 level performance. Balanced inference vs. output length for everyday use. First DeepSeek model with thinking-in-tool-use capability. FP8.

  • Tools
  • Structured output
  • PDF input
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.21
Output
$0.31
Cache read
$0.0216

Reference prices as published by the model provider. API example

Specifications

Context window
163K
Max output
65.5K
Average output (7 days)
408 tokens
Longer than 40% of 465 models

Side by side

DeepSeek V3.2 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 V3.2DeepSeekThis modelDeepSeek LatestDeepSeekRemove
ProviderDeepSeekDeepSeek
Discovery tierOpen weightsOpen weights
AddedDec 1, 2025May 3, 2026
Limits
Context window163K1.05MBest
Max output65.54K384KBest
Average output408 tokens—
Pricing, USD per 1M tokens
Input$0.21Best$0.23
Output$0.31Best$0.69
Cache read$0.0216$0.0073Best
Performance
Throughput46.8 tok/s234.7 tok/sBest
Time to first token3.2 s2.5 sBest
Request success (24h)99.6%99.7%Best
Benchmark indexes
Intelligence Index16.036.0Best
Coding Index—68.8
Math Index59.0—
Capabilities
ReasoningNoYes
ToolsYesYes
Structured outputYesYes
PDF inputYesNo
Reasoning effort—none, low, high, max

Public evaluations

Benchmarks

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

  • DeepSeek V3.2
  • DeepSeek Latest

Intelligence Index

  • DeepSeek V3.216.0
  • DeepSeek Latest36.0

DeepSeek V3.2: Better than 40% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • DeepSeek V3.2: —
    • DeepSeek Latest: 68.8

  • Math Index

    Composite of the math evaluations

    • DeepSeek V3.2: 59.0
    • DeepSeek Latest: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • DeepSeek V3.2: —
    • DeepSeek Latest: 56.7%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • DeepSeek V3.2: —
    • DeepSeek Latest: 25.9%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • DeepSeek V3.2: —
    • DeepSeek Latest: 1,256 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • DeepSeek V3.2: —
    • DeepSeek Latest: 1,449 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • DeepSeek V3.2: —
    • DeepSeek Latest: 11.4%

  • AA-LCRv1.1

    Reasoning across long documents

    • DeepSeek V3.2: 45.7%
    • DeepSeek Latest: 80.3%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • DeepSeek V3.2: 11.2%
    • DeepSeek Latest: 41.0%Top

  • GPQA Diamond

    Graduate-level science questions

    • DeepSeek V3.2: 75.1%
    • DeepSeek Latest: 92.8%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • DeepSeek V3.2: 83.7%
    • DeepSeek Latest: —

  • IFBench

    Following precise output instructions

    • DeepSeek V3.2: 49.0%
    • DeepSeek Latest: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • DeepSeek V3.2: —
    • DeepSeek Latest: 14.1%

  • SciCode

    Scientific computing in Python

    • DeepSeek V3.2: —
    • DeepSeek Latest: 51.0%

  • LiveCodeBench

    Recent competitive programming problems

    • DeepSeek V3.2: 59.3%
    • DeepSeek Latest: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • DeepSeek V3.2: 59.0%
    • DeepSeek Latest: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • DeepSeek V3.2: 32.6%
    • DeepSeek Latest: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • DeepSeek V3.2: 78.9%
    • DeepSeek Latest: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • DeepSeek V3.2: 45.7%
    • DeepSeek Latest: 80.3%Top

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
46.8 tok/s
Faster than 34% of 250 models
Time to first token
3.2 s
Quicker than 53% of 250 models
Prompt cache hits
59.2%
About $0.098 per 1M input tokens with caching
Request success (24h)
99.6%
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 95.6%
  • 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/deepseek-v3.2",
    "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/deepseek-v3.2",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

DeepSeek

Other DeepSeek models, starting with the same family.