NOUMENOpen app
NOUMEN

DeepSeek · Open weights

DeepSeek Chat 0324

Added Jan 1, 2024Open weights

Compare models

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

Loading models…

Use in the app

DeepSeek V3 0324, DeepSeek's 03 March 2025 V3 model, optimized for general-purpose tasks. Quantized at FP8.

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

Pricing

USD per 1M tokens
Input
$0.20
Output
$0.77
Cache read
$0.14

Reference prices as published by the model provider. API example

Specifications

Context window
128K
Max output
8.2K
Average output (7 days)
280 tokens
Longer than 29% of 465 models

Side by side

DeepSeek Chat 0324 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 Chat 0324DeepSeekThis modelDeepSeek V3.1DeepSeekRemove
ProviderDeepSeekDeepSeek
Discovery tierOpen weightsOpen weights
AddedJan 1, 2024Jul 26, 2025
Limits
Context window128K128K
Max output8.19K65.54KBest
Average output280 tokens293 tokens
Pricing, USD per 1M tokens
Input$0.20$0.20
Output$0.77$0.70Best
Cache read$0.14—
Performance
Throughput26.5 tok/sBest14.4 tok/s
Time to first token2.7 s2.1 sBest
Request success (24h)99.9%100.0%Best
Benchmark indexes
Intelligence Index9.713.7Best
Coding Index21.2—
Math Index41.049.7Best
Capabilities
ToolsYesYes
Structured outputYesNo
PDF inputNoYes
Reasoning effort——

Public evaluations

Benchmarks

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

  • DeepSeek Chat 0324
  • DeepSeek V3.1

Intelligence Index

  • DeepSeek Chat 03249.7
  • DeepSeek V3.113.7

DeepSeek Chat 0324: Better than 19% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • DeepSeek Chat 0324: 21.2
    • DeepSeek V3.1: —

  • Math Index

    Composite of the math evaluations

    • DeepSeek Chat 0324: 41.0
    • DeepSeek V3.1: 49.7Top

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • DeepSeek Chat 0324: 0.3%
    • DeepSeek V3.1: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • DeepSeek Chat 0324: 0.0%
    • DeepSeek V3.1: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • DeepSeek Chat 0324: 68 Elo
    • DeepSeek V3.1: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • DeepSeek Chat 0324: 69 Elo
    • DeepSeek V3.1: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • DeepSeek Chat 0324: 1.6%
    • DeepSeek V3.1: —

  • AA-LCRv1.1

    Reasoning across long documents

    • DeepSeek Chat 0324: 40.7%
    • DeepSeek V3.1: 47.0%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • DeepSeek Chat 0324: 4.7%
    • DeepSeek V3.1: 6.7%Top

  • GPQA Diamond

    Graduate-level science questions

    • DeepSeek Chat 0324: 65.5%
    • DeepSeek V3.1: 73.5%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • DeepSeek Chat 0324: 81.9%
    • DeepSeek V3.1: 83.3%Top

  • IFBench

    Following precise output instructions

    • DeepSeek Chat 0324: 41.0%Top
    • DeepSeek V3.1: 37.8%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • DeepSeek Chat 0324: 0.0%
    • DeepSeek V3.1: —

  • SciCode

    Scientific computing in Python

    • DeepSeek Chat 0324: 39.0%
    • DeepSeek V3.1: —

  • LiveCodeBench

    Recent competitive programming problems

    • DeepSeek Chat 0324: 40.5%
    • DeepSeek V3.1: 57.7%Top

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • DeepSeek Chat 0324: 41.0%
    • DeepSeek V3.1: 49.7%Top

  • AIME

    Competition mathematics

    • DeepSeek Chat 0324: 52.0%
    • DeepSeek V3.1: —

  • MATH-500

    Competition math problems

    • DeepSeek Chat 0324: 94.2%
    • DeepSeek V3.1: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • DeepSeek Chat 0324: 15.2%
    • DeepSeek V3.1: 24.2%Top

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • DeepSeek Chat 0324: 47.1%Top
    • DeepSeek V3.1: 34.8%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • DeepSeek Chat 0324: 40.7%
    • DeepSeek V3.1: 47.0%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
26.5 tok/s
Faster than 17% of 251 models
Time to first token
2.7 s
Quicker than 64% of 251 models
Prompt cache hits
69.7%
About $0.15 per 1M input tokens with caching
Request success (24h)
99.9%
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 98.2%
  • 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-v3-0324",
    "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-v3-0324",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

DeepSeek

Other DeepSeek models, starting with the same family.