Gemini 3 Pro Preview (high)

November 18, 2025Last synced Aug 1, 2026, 12:18 AM

Human Frontier

88.3

Human-calibrated frontier signal, backed by Arena-style preference evidence and separate from raw AgMoBench benchmark composite scores.

Blended Price

$4.50/M

Input Price

$2.00/M

Output Price

$12.00/M

Speed

—

TTFT

—

Benchmark Scores

Intelligence Index

Performance Radar

Gemini 3 Pro Preview (high)

Field Average

External Benchmarks

AA-Omniscience Accuracy

How Gemini 3 Pro Preview (high) Compares

Axes

X Axis

Y Axis

Bubble Size

Filters

Blended Price (USD)$0.00 – $37.5

AgMoBench Overall≥ 1.8

Providers

ai2ai21-labsalibabaanthropicarceeawsazurebaidu

Show quadrants

anthropic

openai

mistral

azure

cohere

meta

ai21-labs

inclusionai

google

ai2

ibm

alibaba

upstage

nvidia

aws

kimi

nous-research

thinking-machines

minimax

stepfun

tencent

deepseek

zai

multiversecomputing

xiaomi

xai

deepcogito

sarvam

kwaikat

baidu

swiss-ai-initiative

bytedance_seed

inception

reka-ai

arcee

perplexity

Bubble size = Context Window

8191.00

2000000.00

Compare with other models

Terminal-Bench 2.1terminal_bench

v2.1 · official-leaderboard · Gemini 3 Pro (high) · harness gemini-cli@0.40.0 · effort high · 5 attempts · harbor:terminal-bench/terminal-bench-2-1 · provider gemini/gemini-3-pro-preview · record leaderboard/submissions/2026-05-01-gemini-gemini-3-pro-preview-high-gemini-cli.json · run fd8707bb-51e8-56fa-8e46-769a82a531ae

View run source

Source metadata

{
  "passAt2": 0.7427,
  "passAt3": 0.7831,
  "passAt4": 0.8067,
  "passAt5": 0.8202,
  "taskCount": 89,
  "actualTokens": {
    "output": 7219388,
    "cachedInput": 329123284,
    "uncachedInput": 47653156
  },
  "datasetPackage": "terminal-bench/terminal-bench-2-1",
  "leaderboardUrl": "https://www.tbench.ai/leaderboard/terminal-bench/2.1",
  "pullRequestUrl": "https://github.com/harbor-framework/terminal-bench-2-1/pull/66",
  "agentDisplayUrl": "https://github.com/google-gemini/gemini-cli",
  "modelDisplayUrl": "https://blog.google/products-and-platforms/products/gemini/gemini-3/",
  "actualTrialCount": 445,
  "agentOrganization": "Google",
  "modelOrganization": "Google",
  "actualTotalCostUsd": 247.76,
  "rewardHacksPercent": 0.45,
  "accuracyStandardError": 1.38,
  "submissionDownloadUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench-2-1/main/leaderboard/submissions/2026-05-01-gemini-gemini-3-pro-preview-high-gemini-cli.json",
  "disqualifiedTrialCount": 2,
  "expectedAttemptsPerTask": 5,
  "sourceFilterReasoningEffort": "high",
  "actualAverageTrialDurationSeconds": 436.6
}

Evidence observed 2026-05-01

65.8 / 100

Terminal-Bench 2.1terminal_bench

v2.1 · official-leaderboard · Gemini 3 Pro (high) · harness terminus-2@2.0.0 · effort high · 5 attempts · harbor:terminal-bench/terminal-bench-2-1 · provider gemini/gemini-3-pro-preview · record leaderboard/submissions/2026-05-01-gemini-gemini-3-pro-preview-high-terminus-2.json · run 10e2e56b-ed31-5f65-a489-69f78b902adf

View run source

Source metadata

{
  "passAt2": 0.8135,
  "passAt3": 0.8472,
  "passAt4": 0.8652,
  "passAt5": 0.8764,
  "taskCount": 89,
  "actualTokens": {
    "output": 12862730,
    "cachedInput": 89495525,
    "uncachedInput": 26094316
  },
  "datasetPackage": "terminal-bench/terminal-bench-2-1",
  "leaderboardUrl": "https://www.tbench.ai/leaderboard/terminal-bench/2.1",
  "pullRequestUrl": "https://github.com/harbor-framework/terminal-bench-2-1/pull/48",
  "agentDisplayUrl": "https://www.tbench.ai/news/terminus",
  "modelDisplayUrl": "https://blog.google/products-and-platforms/products/gemini/gemini-3/",
  "actualTrialCount": 445,
  "agentOrganization": "Terminal-Bench",
  "modelOrganization": "Google",
  "actualTotalCostUsd": 224.44,
  "rewardHacksPercent": 0.45,
  "accuracyStandardError": 1.29,
  "submissionDownloadUrl": "https://raw.githubusercontent.com/harbor-framework/terminal-bench-2-1/main/leaderboard/submissions/2026-05-01-gemini-gemini-3-pro-preview-high-terminus-2.json",
  "disqualifiedTrialCount": 2,
  "expectedAttemptsPerTask": 5,
  "sourceFilterReasoningEffort": "high",
  "actualAverageTrialDurationSeconds": 547.8
}

Evidence observed 2026-05-01

73.9 / 100