NOUMENOpen app
NOUMEN

Google · Efficient

Gemini 3 Flash Thinking

Added Dec 17, 2025

Compare models

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

Loading models…

Use in the app

Google's Gemini 3 Flash preview model with thinking mode enabled for enhanced reasoning and chain-of-thought capabilities. Audio input costs $1.00 per million audio tokens.

  • Reasoning
  • Vision
  • Audio input
  • Video input
Input
Text, Images, Video, Audio
Output
Text
Reasoning effort
minimal · low · medium · high
Discovery tier
Efficient

Pricing

USD per 1M tokens
Input
$0.50
Output
$3.00
Cache read
$0.05

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
65.5K
Average output (7 days)
1.7K tokens
Longer than 89% of 468 models

Side by side

Gemini 3 Flash Thinking 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…

Gemini 3 Flash ThinkingGoogleThis modelClaude 4.6 OpusAnthropicRemove
ProviderGoogleAnthropic
Discovery tierEfficientFrontier
AddedDec 17, 2025Feb 5, 2026
Limits
Context window1.05MBest1M
Max output65.54K128KBest
Average output1.7K tokens1.4K tokens
Pricing, USD per 1M tokens
Input$0.50Best$5.00
Output$3.00Best$25.00
Cache read$0.05Best$0.50
Performance
Throughput153.3 tok/sBest40.5 tok/s
Time to first token6.4 s3.5 sBest
Request success (24h)100.0%100.0%
Benchmark indexes
Intelligence Index26.326.4Best
Math Index97.0—
Capabilities
ReasoningYesYes
ToolsNoYes
VisionYesYes
Structured outputNoYes
PDF inputNoYes
Audio inputYesNo
Video inputYesNo
Reasoning effortminimal, low, medium, highlow, medium, high, xhigh, max

Public evaluations

Benchmarks

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

  • Gemini 3 Flash Thinking
  • Claude 4.6 Opus

Intelligence Index

  • Gemini 3 Flash Thinking26.3
  • Claude 4.6 Opus26.4

Gemini 3 Flash Thinking: Better than 66% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • Gemini 3 Flash Thinking: 97.0
    • Claude 4.6 Opus: —

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • Gemini 3 Flash Thinking: 78.0%Top
    • Claude 4.6 Opus: 67.0%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Gemini 3 Flash Thinking: 36.6%Top
    • Claude 4.6 Opus: 19.1%

  • GPQA Diamond

    Graduate-level science questions

    • Gemini 3 Flash Thinking: 89.8%Top
    • Claude 4.6 Opus: 84.0%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Gemini 3 Flash Thinking: 89.0%
    • Claude 4.6 Opus: —

  • IFBench

    Following precise output instructions

    • Gemini 3 Flash Thinking: 78.0%Top
    • Claude 4.6 Opus: 44.6%

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • Gemini 3 Flash Thinking: 90.8%
    • Claude 4.6 Opus: —

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Gemini 3 Flash Thinking: 97.0%
    • Claude 4.6 Opus: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Gemini 3 Flash Thinking: 38.6%
    • Claude 4.6 Opus: 48.5%Top

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Gemini 3 Flash Thinking: 80.4%
    • Claude 4.6 Opus: 84.8%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Gemini 3 Flash Thinking: 78.0%Top
    • Claude 4.6 Opus: 67.0%

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
153.3 tok/s
Faster than 76% of 254 models
Time to first token
6.4 s
Quicker than 27% of 254 models
Prompt cache hits
29.9%
About $0.37 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 80.0%
  • 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": "google/gemini-3-flash-preview-thinking",
    "reasoning_effort": "medium",
    "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="google/gemini-3-flash-preview-thinking",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.