Skip to content
aiqbench

Google DeepMind · model passport

Google: Gemini 3 Flash Preview

gemini-3-flash-preview
  • 1,048,576 context
13/27axes covered
Best percentile
99.6LiveCodeBench (Artificial Analysis)
Composite standing
72.1shrunk · 13/27 axes

What the data says

Benchmarked on 13 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 99.6 pct — LiveCodeBench (Artificial Analysis) (90.8%)
  • 99.3 pct — MMLU-Pro (Artificial Analysis) (89.0%)
  • 96.6 pct — IFBench (Artificial Analysis) (78.0%)

Uneven: 81.5 percentile points between its best and worst axis — 99.6 pct on LiveCodeBench (Artificial Analysis) versus 18.1 pct on HHEM Hallucination (Vectara).

  • LiveCodeBench (Artificial Analysis)99.6 pct90.8%Verify →
  • MMLU-Pro (Artificial Analysis)99.3 pct89.0%Verify →
  • IFBench (Artificial Analysis)96.6 pct78.0%Verify →
  • LiveBench89.3 pct73.7%Verify →
  • SimpleQA Verified87.8 pct66.8%Verify →
  • AIME (math)81.5 pct95.6%Verify →
  • FrontierMath81.2 pct35.6%Verify →
  • Terminal-Bench79.5 pct64.3%Verify →
  • GPQA Diamond79.2 pct89.4%Verify →
  • SimpleBench74.7 pct61.1%Verify →
  • SWE-bench Verified62.5 pct75.4%Verify →
  • ARC-AGI-255.1 pct33.6%Verify →
  • HHEM Hallucination (Vectara)18.1 pct86.5%Verify →
  • OverallNot benchmarked
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Factuality ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • Humanity's Last ExamNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • CodingNot benchmarked
  • Instruction followingNot benchmarked
  • Hard promptsNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Use it hosted

API model IDgoogle/gemini-3-flash-preview
Context length1,048,576 tokens
Max output65,536 tokens
Acceptstext, image, file, audio, video
Input price$0.50 per 1M tokens
Cached input price$0.05 per 1M tokens
Output price$3 per 1M tokens

API parameters this endpoint supports

Include reasoningMax tokensReasoningReasoning effortResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"google/gemini-3-flash-preview","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • gemini-3-flash-preview stands at 72.1 (raw mean 77.3) — measured on 13 of 27 axes.

    standingtemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →