NOUMENOpen app
NOUMEN

inclusionai · Efficient

Ling 3.0 Flash

Added Jul 23, 2026

Compare models

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

Loading models…

Use in the app

Ling-3.0-flash is a 124B-parameter Mixture-of-Experts model with approximately 5.1B parameters active per token. It prioritizes token efficiency and production-scale agentic inference, helping coding and tool-using agents complete more work within constrained latency and serving budgets.

  • Tools
Input
Text
Output
Text
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.075
Output
$0.22
Cache read
$0.015

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
32.8K
Average output (7 days)
1.1K tokens
Longer than 75% of 468 models

Side by side

Ling 3.0 Flash 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…

Ling 3.0 FlashinclusionaiThis modelLing 3.1 FlashinclusionaiRemove
Providerinclusionaiinclusionai
Discovery tierEfficientOpen weights
AddedJul 23, 2026Sep 30, 2026
Limits
Context window262.14K262.14K
Max output32.77K32.77K
Average output1.1K tokens7.8K tokens
Pricing, USD per 1M tokens
Input$0.075$0.075
Output$0.22$0.22
Cache read$0.015$0.015
Performance
Throughput279.2 tok/sBest94.2 tok/s
Time to first token1.7 sBest4.5 s
Request success (24h)99.8%100.0%Best
Benchmark indexes
Intelligence Index20.141.1Best
Coding Index50.6—
Capabilities
ReasoningNoYes
ToolsYesYes
Structured outputNoYes
Reasoning effort—none, low, medium, high

Public evaluations

Benchmarks

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

  • Ling 3.0 Flash
  • Ling 3.1 Flash

Intelligence Index

  • Ling 3.0 Flash20.1
  • Ling 3.1 Flash41.1

Ling 3.0 Flash: Better than 49% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Ling 3.0 Flash: 50.6
    • Ling 3.1 Flash: —

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Ling 3.0 Flash: 3.2%
    • Ling 3.1 Flash: 61.7%Top

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Ling 3.0 Flash: 0.6%
    • Ling 3.1 Flash: 23.6%Top

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Ling 3.0 Flash: 798 Elo
    • Ling 3.1 Flash: 1,397 EloTop

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Ling 3.0 Flash: 948 Elo
    • Ling 3.1 Flash: 1,617 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Ling 3.0 Flash: 5.4%
    • Ling 3.1 Flash: 10.0%Top

  • AA-LCRv1.1

    Reasoning across long documents

    • Ling 3.0 Flash: 73.0%
    • Ling 3.1 Flash: 83.0%Top

  • MLCR-AA

    Long-context reasoning over medical records

    • Ling 3.0 Flash: 5.0%
    • Ling 3.1 Flash: 12.2%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Ling 3.0 Flash: 23.7%
    • Ling 3.1 Flash: 39.4%Top

  • GPQA Diamond

    Graduate-level science questions

    • Ling 3.0 Flash: 85.5%
    • Ling 3.1 Flash: —

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Ling 3.0 Flash: 0.0%
    • Ling 3.1 Flash: 33.3%Top

  • SciCode

    Scientific computing in Python

    • Ling 3.0 Flash: 42.0%
    • Ling 3.1 Flash: 54.1%Top

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Ling 3.0 Flash: 73.0%
    • Ling 3.1 Flash: 83.0%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
279.2 tok/s
Faster than 92% of 254 models
Time to first token
1.7 s
Quicker than 87% of 254 models
Prompt cache hits
77.0%
About $0.0288 per 1M input tokens with caching
Request success (24h)
99.8%
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 33.3%
  • 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": "inclusionai/ling-3.0-flash",
    "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="inclusionai/ling-3.0-flash",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

inclusionai

Other inclusionai models, starting with the same family.