NOUMENOpen app
NOUMEN

aionlabs · Efficient

AionLabs: Aion 3.0 Mini

Added Jul 7, 2026

Compare models

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

Loading models…

Use in the app

Aion 3.0 Mini is a DeepSeek-family collaborative generation model tuned for immersive roleplay and storytelling, with stronger narrative structure, tension, conflict, and nuanced mature themes.

  • Reasoning
  • Tools
Input
Text
Output
Text
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.70
Output
$1.40
Cache read
$0.18

Reference prices as published by the model provider. API example

Specifications

Context window
131.1K
Max output
32.8K
Average output (7 days)
1K tokens
Longer than 71% of 473 models

Side by side

AionLabs: Aion 3.0 Mini 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…

AionLabs: Aion 3.0 MiniaionlabsThis modelGemini 3.8 FlashGoogleRemove
ProvideraionlabsGoogle
Discovery tierEfficientEfficient
AddedJul 7, 2026Sep 2, 2026
Limits
Context window131.07K1.05MBest
Max output32.77K65.54KBest
Average output1K tokens1.2K tokens
Pricing, USD per 1M tokens
Input$0.70Best$0.75
Output$1.40Best$3.75
Cache read$0.18$0.075Best
Cache write—$0.0417
Performance
Throughput61.4 tok/s244.5 tok/sBest
Time to first token3.0 sBest8.5 s
Request success (24h)100.0%Best99.8%
Benchmark indexes
Intelligence Index—39.8
Coding Index—74.1
Capabilities
ReasoningYesYes
ToolsYesYes
VisionNoYes
Structured outputNoYes
PDF inputNoYes
Audio inputNoYes
Video inputNoYes
Reasoning effort—minimal, low, medium, high

Public evaluations

Benchmarks

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

  • AionLabs: Aion 3.0 Mini
  • Gemini 3.8 Flash

Intelligence Index

  • AionLabs: Aion 3.0 Mini—
  • Gemini 3.8 Flash39.8
3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 74.1

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 60.9%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 31.1%

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 1,181 Elo

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 1,423 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 22.8%

  • AA-LCRv1.1

    Reasoning across long documents

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 84.0%

  • MLCR-AA

    Long-context reasoning over medical records

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 12.8%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 42.1%

  • GPQA Diamond

    Graduate-level science questions

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 93.5%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 19.7%

  • SciCode

    Scientific computing in Python

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 55.1%

Legacy evaluations

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • AionLabs: Aion 3.0 Mini: —
    • Gemini 3.8 Flash: 84.0%

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
61.4 tok/s
Faster than 43% of 260 models
Time to first token
3.0 s
Quicker than 52% of 260 models
Prompt cache hits
46.0%
About $0.46 per 1M input tokens with caching
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": "aion-labs/aion-3.0-mini",
    "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="aion-labs/aion-3.0-mini",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

aionlabs

Other aionlabs models, starting with the same family.