NOUMENOpen app
NOUMEN

DeepSeek · Open weights

DeepSeek V4 Flash 0731

Added Jul 31, 2026Open weights

Compare models

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

Loading models…

Use in the app

DeepSeek V4 Flash 0731 is a re-post-trained Mixture-of-Experts model with a 1M-token context window, built for coding, reasoning, and agent workflows.

  • Reasoning
  • Tools
  • Structured output
Input
Text
Output
Text
Reasoning effort
none · low · high · max
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.0444
Output
$0.13
Cache read
$0.0014

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
131.1K
Average output (7 days)
451 tokens
Longer than 44% of 468 models

Side by side

DeepSeek V4 Flash 0731 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 V4 Flash 0731DeepSeekThis modelDeepSeek V4.1 FlashDeepSeekRemove
ProviderDeepSeekDeepSeek
Discovery tierOpen weightsOpen weights
AddedJul 31, 2026Sep 8, 2026
Limits
Context window1.05MBest1M
Max output131.07K384KBest
Average output451 tokens376 tokens
Pricing, USD per 1M tokens
Input$0.0444Best$0.09
Output$0.13Best$0.25
Cache read$0.0014Best$0.006
Performance
Throughput89.0 tok/s146.8 tok/sBest
Time to first token3.4 s1.9 sBest
Request success (24h)99.9%Best99.8%
Benchmark indexes
Intelligence Index34.3Best24.7
Coding Index69.1—
Capabilities
ReasoningYesYes
ToolsYesYes
VisionNoYes
Structured outputYesYes
Reasoning effortnone, low, high, maxnone, low, high, max

Public evaluations

Benchmarks

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

  • DeepSeek V4 Flash 0731
  • DeepSeek V4.1 Flash

Intelligence Index

  • DeepSeek V4 Flash 073134.3
  • DeepSeek V4.1 Flash24.7

DeepSeek V4 Flash 0731 tested at max reasoning effort

DeepSeek V4 Flash 0731: Better than 78% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • DeepSeek V4 Flash 0731: 69.1
    • DeepSeek V4.1 Flash: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • DeepSeek V4 Flash 0731: 54.0%Top
    • DeepSeek V4.1 Flash: 47.8%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • DeepSeek V4 Flash 0731: 23.0%Top
    • DeepSeek V4.1 Flash: 14.2%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • DeepSeek V4 Flash 0731: 1,257 EloTop
    • DeepSeek V4.1 Flash: 1,118 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • DeepSeek V4 Flash 0731: 1,433 EloTop
    • DeepSeek V4.1 Flash: 1,340 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • DeepSeek V4 Flash 0731: 11.0%Top
    • DeepSeek V4.1 Flash: 3.8%

  • AA-LCRv1.1

    Reasoning across long documents

    • DeepSeek V4 Flash 0731: 79.7%Top
    • DeepSeek V4.1 Flash: 55.3%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • DeepSeek V4 Flash 0731: 38.6%Top
    • DeepSeek V4.1 Flash: 10.8%

  • GPQA Diamond

    Graduate-level science questions

    • DeepSeek V4 Flash 0731: 90.8%
    • DeepSeek V4.1 Flash: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • DeepSeek V4 Flash 0731: 12.1%Top
    • DeepSeek V4.1 Flash: 5.6%

  • SciCode

    Scientific computing in Python

    • DeepSeek V4 Flash 0731: 50.3%Top
    • DeepSeek V4.1 Flash: 35.5%

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • DeepSeek V4 Flash 0731: 79.7%Top
    • DeepSeek V4.1 Flash: 55.3%

Scores from Artificial Analysis, updated Oct 26, 2026. DeepSeek V4 Flash 0731 was tested at max reasoning effort; compare scores measured at the same effort.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
89.0 tok/s
Faster than 58% of 254 models
Time to first token
3.4 s
Quicker than 51% of 254 models
Prompt cache hits
82.4%
About $0.009 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 99.4%
  • 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-v4-flash-0731",
    "reasoning_effort": "none",
    "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-v4-flash-0731",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

DeepSeek

Other DeepSeek models, starting with the same family.