THE SCOREBOARD FOR WHAT'S NEXTAI + QUANTUM, SCORED. COMPARED. EXPLAINED.
AIQUANTUMSCORE.

Google / Gemini

Gemini 3.8 Flash

Version gemini-3-8-flash · Released 2026-09-02 · Available

Compare this model ↗
57QUANTUM SCORE

Explore Google history →

INDEPENDENT BENCHMARK / SNAPSHOT 2026-10-16

Independent benchmark snapshot

This snapshot is over 14 days old. Review the dated benchmark records below for newer evidence.

Artificial Analysis Intelligence Index v4.3.2 · Higher is better. Selected models, sorted by this benchmark’s published score; equal scores are ties.

ModelAA Intelligence IndexTested effortEvidence
Gemini 3.8 Flash41HighIndependent result ↗

This is a dated snapshot from one independent evaluator, on its own scale—not a Quantum Score or a complete market ranking. A missing Quantum Score means our broader evidence requirements are unmet; it does not mean the model is worse. Test dates are not published in these source summaries.

THE PRACTICAL READ

Should you shortlist Gemini 3.8 Flash?

Evidence checked 2026-11-07. This interpretation uses the cited records below; it is not a hands-on review.

What the evidence supports

The strongest relative category results in our current evidence are math and multimodal. Scores describe this comparison pool, not a percentage of real-world tasks solved.

16 current independent benchmark observations. Several tasks from one suite do not count as separate independent sources.

What remains uncertain

Category coverage does not establish reliability on your own workload. Writing quality, factual accuracy, privacy and production reliability require separate evaluation.

An overall score summarizes covered categories; a high overall score may not match your priorities. Check the effort level and tool configuration before comparing results.

Inspect coverage →

Cost and response time

At 10M input + 2M output tokens, the base API estimate is $15.00/month. This excludes tools, caching, batch rates, subscriptions and taxes.

Output tokens per second measures generation after it starts. Reasoning time can make a fast-generating model take longer to finish. Use the measured speed and effort records below.

Estimate your actual usage →

Alternatives worth comparing

Grok 4.6

14 shared independent observations with matching test methods.

Same illustrative usage: $32.00/month. Lower price does not establish equivalent task performance.

Compare tradeoffs →

GPT-6 Astra

14 shared independent observations with matching test methods.

Same illustrative usage: $200.00/month. Lower price does not establish equivalent task performance.

Compare tradeoffs →

GPT-5.6 Terra

13 shared independent observations with matching test methods.

Same illustrative usage: $44.00/month. Lower price does not establish equivalent task performance.

Compare tradeoffs →

Follow changes to this model ↗ Choose an AI for coding → · Research guide → · Automation guide →

Coding32
Reasoning61
Math95
REVAL™
ⓘREVAL™ Research Evaluation Index — our proprietary 0–100 benchmark measuring real-world AI research performance. Methodology →
Not tested
Agents30
Multimodal91
SpeedNot measured
Value54

PROPRIETARY / RESEARCH EVALUATION INDEX

REVAL™ Not tested

Methodology →

This exact model version has not completed a reviewed REVAL evaluation. No score is assigned from reputation or external benchmarks.

01 / EVIDENCE

Benchmark breakdown

BenchmarkVersionMethodConfigurationResultTest dateVerifiedSource typeSource
ARC-AGI-3 Semi-Private Standard harness best effort3 Semi-PrivateARC Prize Standard harness; best published effort per model—10.3663 %2026-09-292026-11-07benchmark maintainerView source ↗
Humanity's Last Exam AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High)AA Index 4.3.2Gemini High—48 %2026-09-292026-09-29independent labView source ↗
AutomationBench-AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High)AA Index 4.3.2Gemini High—60 %2026-09-292026-09-29independent labView source ↗
LiveBench code_completion2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high71.739 %not published2026-10-17benchmark maintainerView source ↗
LiveBench code_generation2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high73.239 %not published2026-10-17benchmark maintainerView source ↗
LiveBench javascript2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high72.727 %not published2026-10-17benchmark maintainerView source ↗
LiveBench AMPS_Hard2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high99 %not published2026-10-17benchmark maintainerView source ↗
LiveBench olympiad2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high92.159 %not published2026-10-17benchmark maintainerView source ↗
LiveBench theory_of_mind2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high76.923 %not published2026-10-17benchmark maintainerView source ↗
LiveBench zebra_puzzle2026-06-25Public table; published model effort variantLiveBench public table 2026-06-25 · gemini-3.8-flash-high98.25 %not published2026-10-17benchmark maintainerView source ↗
MedXpertQA MM Mercor public leaderboardMedXpertQA MM Mercor 200 public tasksMercor independent evaluation200 public tasks; Pass@1; model effort recorded separately · High84.7 %not published2026-09-30independent labView source ↗
DeepSWE v1.1 maintainer pass@11.1mini-swe-agent; 113-task set; best published reasoning configuration per model; provider/verifier/network errors excluded as documentedmini-swe-agent; 113 tasks; best published reasoning configuration · high73.825503 %not published2026-10-01benchmark maintainerView source ↗
Terminal-Bench 4.0 Vals mini-SWE-agent avg@3 fallbacks failed4.0Vals mini-SWE-agent avg@3; Anthropic fallbacks treated as failures— · high19.192 %not published2026-10-01independent labView source ↗
LiveBench 2.0 coding20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in table82.2 %not published2026-11-07benchmark maintainerView source ↗
LiveBench 2.0 reasoning20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in table89.3 %not published2026-11-07benchmark maintainerView source ↗
LiveBench 2.0 long_agentic20261025LiveBench 2.0 published release tablePublished model configuration; effort not specified in table16.5 %not published2026-11-07benchmark maintainerView source ↗

02 / COST

Price & speed

Input / 1M tokens $0.75

Output / 1M tokens $3.75

Cached input / 1M —

Batch input / 1M —

Batch output / 1M —

Speed No comparable observation yet

Price verified 2026-09-29

Price scope Introductory rate through 2026-12-31

Price source Official pricing ↗

02A / PERFORMANCE

Measured speed

No sourced speed measurement at a stated effort level yet.

03 / PROFILE

Capabilities

visiontool use

04 / CONTEXT

Evidence coverage & limits

Published tests describe specific environments and effort settings. They do not establish performance on every task. Read the practical interpretation above →

Coverage: 85% · Moderate confidence · 16 observations · 7 source domains (7 independent).

05 / HISTORY

Version & price history

Version gemini-3-8-flash · released 2026-09-02.

2026-09-29: $0.75 input / $3.75 output per 1M tokens · Price source ↗

06 / SOURCES

Source records

DATED EVIDENCE

History

Standard API price history

2026-09-292026-09-29
  • 2026-09-29: 1.5 USD / 1M · 75% input / 25% output USD per 1M · source ↗

One observation; no trend can be inferred.

ARC-AGI-3 Semi-Private Standard harness best effort / 3 Semi-Private / %

2026-09-292026-09-29
  • 2026-09-29: 10.366 result · ARC Prize Standard harness; best published effort per model · source ↗

One observation; no trend can be inferred.

Humanity's Last Exam AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High) / AA Index 4.3.2 / %

2026-09-292026-09-29
  • 2026-09-29: 48 result · Gemini High · source ↗

One observation; no trend can be inferred.

AutomationBench-AA Index v4.3.2 (Opus Max fallback, Astra Max, Gemini High) / AA Index 4.3.2 / %

2026-09-292026-09-29
  • 2026-09-29: 60 result · Gemini High · source ↗

One observation; no trend can be inferred.

LiveBench code_completion / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 71.739 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench code_generation / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 73.239 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench javascript / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 72.727 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench AMPS_Hard / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 99 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench olympiad / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 92.159 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench theory_of_mind / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 76.923 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench zebra_puzzle / 2026-06-25 / %

2026-10-172026-10-17
  • 2026-10-17: 98.25 result · Public table; published model effort variant · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

MedXpertQA MM Mercor public leaderboard / MedXpertQA MM Mercor 200 public tasks / %

2026-09-302026-09-30
  • 2026-09-30: 84.7 result · Mercor independent evaluation · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

DeepSWE v1.1 maintainer pass@1 / 1.1 / %

2026-10-012026-10-01
  • 2026-10-01: 73.826 result · mini-swe-agent; 113-task set; best published reasoning configuration per model; provider/verifier/network errors excluded as documented · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

Terminal-Bench 4.0 Vals mini-SWE-agent avg@3 fallbacks failed / 4.0 / %

2026-10-012026-10-01
  • 2026-10-01: 19.192 result · Vals mini-SWE-agent avg@3; Anthropic fallbacks treated as failures · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench 2.0 coding / 20261025 / %

2026-11-072026-11-07
  • 2026-11-07: 82.2 result · LiveBench 2.0 published release table · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench 2.0 reasoning / 20261025 / %

2026-11-072026-11-07
  • 2026-11-07: 89.3 result · LiveBench 2.0 published release table · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

LiveBench 2.0 long_agentic / 20261025 / %

2026-11-072026-11-07
  • 2026-11-07: 16.5 result · LiveBench 2.0 published release table · evaluation date not published; verification date shown · source ↗

One observation; no trend can be inferred.

coding category score history

2026-10-012026-11-07
  • 2026-10-01: 16 score · Scoring 3.0.0
  • 2026-10-02: 40 score · Scoring 3.0.0
  • 2026-10-03: 40 score · Scoring 3.0.0
  • 2026-10-04: 40 score · Scoring 3.0.0
  • 2026-10-05: 40 score · Scoring 3.0.0
  • 2026-10-06: 40 score · Scoring 3.0.0
  • 2026-10-07: 40 score · Scoring 3.0.0
  • 2026-10-08: 40 score · Scoring 3.0.0
  • 2026-10-09: 40 score · Scoring 3.0.0
  • 2026-10-10: 40 score · Scoring 3.0.0
  • 2026-10-11: 40 score · Scoring 3.0.0
  • 2026-10-12: 40 score · Scoring 3.0.0
  • 2026-10-13: 40 score · Scoring 3.0.0
  • 2026-10-14: 40 score · Scoring 3.0.0
  • 2026-10-15: 40 score · Scoring 3.0.0
  • 2026-10-16: 40 score · Scoring 3.0.0
  • 2026-10-17: 40 score · Scoring 3.0.0
  • 2026-10-18: 40 score · Scoring 3.0.0
  • 2026-10-19: 40 score · Scoring 3.0.0
  • 2026-10-20: 40 score · Scoring 3.0.0
  • 2026-10-21: 40 score · Scoring 3.0.0
  • 2026-10-22: 40 score · Scoring 3.0.0
  • 2026-10-23: 40 score · Scoring 3.0.0
  • 2026-10-24: 40 score · Scoring 3.0.0
  • 2026-10-25: 40 score · Scoring 3.0.0
  • 2026-10-26: 40 score · Scoring 3.0.0
  • 2026-10-27: 40 score · Scoring 3.0.0
  • 2026-10-28: 40 score · Scoring 3.0.0
  • 2026-10-29: 40 score · Scoring 3.0.0
  • 2026-10-30: 40 score · Scoring 3.0.0
  • 2026-10-31: 40 score · Scoring 3.0.0
  • 2026-11-01: 40 score · Scoring 3.0.0
  • 2026-11-02: 40 score · Scoring 3.0.0
  • 2026-11-03: 40 score · Scoring 3.0.0
  • 2026-11-04: 40 score · Scoring 3.0.0
  • 2026-11-05: 40 score · Scoring 3.0.0
  • 2026-11-06: 40 score · Scoring 3.0.0
  • 2026-11-07: 40 score · Scoring 3.0.0

reasoning category score history

2026-09-302026-11-07
  • 2026-09-30: 8 score · Scoring 3.0.0
  • 2026-10-01: 62 score · Scoring 3.0.0
  • 2026-10-02: 62 score · Scoring 3.0.0
  • 2026-10-03: 62 score · Scoring 3.0.0
  • 2026-10-04: 62 score · Scoring 3.0.0
  • 2026-10-05: 62 score · Scoring 3.0.0
  • 2026-10-06: 62 score · Scoring 3.0.0
  • 2026-10-07: 62 score · Scoring 3.0.0
  • 2026-10-08: 62 score · Scoring 3.0.0
  • 2026-10-09: 62 score · Scoring 3.0.0
  • 2026-10-10: 62 score · Scoring 3.0.0
  • 2026-10-11: 62 score · Scoring 3.0.0
  • 2026-10-12: 62 score · Scoring 3.0.0
  • 2026-10-13: 62 score · Scoring 3.0.0
  • 2026-10-14: 62 score · Scoring 3.0.0
  • 2026-10-15: 62 score · Scoring 3.0.0
  • 2026-10-16: 62 score · Scoring 3.0.0
  • 2026-10-17: 62 score · Scoring 3.0.0
  • 2026-10-18: 62 score · Scoring 3.0.0
  • 2026-10-19: 62 score · Scoring 3.0.0
  • 2026-10-20: 62 score · Scoring 3.0.0
  • 2026-10-21: 62 score · Scoring 3.0.0
  • 2026-10-22: 62 score · Scoring 3.0.0
  • 2026-10-23: 62 score · Scoring 3.0.0
  • 2026-10-24: 62 score · Scoring 3.0.0
  • 2026-10-25: 62 score · Scoring 3.0.0
  • 2026-10-26: 62 score · Scoring 3.0.0
  • 2026-10-27: 62 score · Scoring 3.0.0
  • 2026-10-28: 62 score · Scoring 3.0.0
  • 2026-10-29: 62 score · Scoring 3.0.0
  • 2026-10-30: 62 score · Scoring 3.0.0
  • 2026-10-31: 62 score · Scoring 3.0.0
  • 2026-11-01: 62 score · Scoring 3.0.0
  • 2026-11-02: 62 score · Scoring 3.0.0
  • 2026-11-03: 62 score · Scoring 3.0.0
  • 2026-11-04: 62 score · Scoring 3.0.0
  • 2026-11-05: 62 score · Scoring 3.0.0
  • 2026-11-06: 62 score · Scoring 3.0.0
  • 2026-11-07: 62 score · Scoring 3.0.0

math category score history

2026-10-012026-11-07
  • 2026-10-01: 95 score · Scoring 3.0.0
  • 2026-10-02: 95 score · Scoring 3.0.0
  • 2026-10-03: 95 score · Scoring 3.0.0
  • 2026-10-04: 95 score · Scoring 3.0.0
  • 2026-10-05: 95 score · Scoring 3.0.0
  • 2026-10-06: 95 score · Scoring 3.0.0
  • 2026-10-07: 95 score · Scoring 3.0.0
  • 2026-10-08: 95 score · Scoring 3.0.0
  • 2026-10-09: 95 score · Scoring 3.0.0
  • 2026-10-10: 95 score · Scoring 3.0.0
  • 2026-10-11: 95 score · Scoring 3.0.0
  • 2026-10-12: 95 score · Scoring 3.0.0
  • 2026-10-13: 95 score · Scoring 3.0.0
  • 2026-10-14: 95 score · Scoring 3.0.0
  • 2026-10-15: 95 score · Scoring 3.0.0
  • 2026-10-16: 95 score · Scoring 3.0.0
  • 2026-10-17: 95 score · Scoring 3.0.0
  • 2026-10-18: 95 score · Scoring 3.0.0
  • 2026-10-19: 95 score · Scoring 3.0.0
  • 2026-10-20: 95 score · Scoring 3.0.0
  • 2026-10-21: 95 score · Scoring 3.0.0
  • 2026-10-22: 95 score · Scoring 3.0.0
  • 2026-10-23: 95 score · Scoring 3.0.0
  • 2026-10-24: 95 score · Scoring 3.0.0
  • 2026-10-25: 95 score · Scoring 3.0.0
  • 2026-10-26: 95 score · Scoring 3.0.0
  • 2026-10-27: 95 score · Scoring 3.0.0
  • 2026-10-28: 95 score · Scoring 3.0.0
  • 2026-10-29: 95 score · Scoring 3.0.0
  • 2026-10-30: 95 score · Scoring 3.0.0
  • 2026-10-31: 95 score · Scoring 3.0.0
  • 2026-11-01: 95 score · Scoring 3.0.0
  • 2026-11-02: 95 score · Scoring 3.0.0
  • 2026-11-03: 95 score · Scoring 3.0.0
  • 2026-11-04: 95 score · Scoring 3.0.0
  • 2026-11-05: 95 score · Scoring 3.0.0
  • 2026-11-06: 95 score · Scoring 3.0.0
  • 2026-11-07: 95 score · Scoring 3.0.0

agents category score history

2026-09-302026-11-07
  • 2026-09-30: 0 score · Scoring 3.0.0
  • 2026-10-01: 60 score · Scoring 3.0.0
  • 2026-10-02: 60 score · Scoring 3.0.0
  • 2026-10-03: 60 score · Scoring 3.0.0
  • 2026-10-04: 60 score · Scoring 3.0.0
  • 2026-10-05: 60 score · Scoring 3.0.0
  • 2026-10-06: 60 score · Scoring 3.0.0
  • 2026-10-07: 60 score · Scoring 3.0.0
  • 2026-10-08: 60 score · Scoring 3.0.0
  • 2026-10-09: 60 score · Scoring 3.0.0
  • 2026-10-10: 60 score · Scoring 3.0.0
  • 2026-10-11: 60 score · Scoring 3.0.0
  • 2026-10-12: 60 score · Scoring 3.0.0
  • 2026-10-13: 60 score · Scoring 3.0.0
  • 2026-10-14: 60 score · Scoring 3.0.0
  • 2026-10-15: 60 score · Scoring 3.0.0
  • 2026-10-16: 60 score · Scoring 3.0.0
  • 2026-10-17: 60 score · Scoring 3.0.0
  • 2026-10-18: 60 score · Scoring 3.0.0
  • 2026-10-19: 60 score · Scoring 3.0.0
  • 2026-10-20: 60 score · Scoring 3.0.0
  • 2026-10-21: 60 score · Scoring 3.0.0
  • 2026-10-22: 60 score · Scoring 3.0.0
  • 2026-10-23: 60 score · Scoring 3.0.0
  • 2026-10-24: 60 score · Scoring 3.0.0
  • 2026-10-25: 60 score · Scoring 3.0.0
  • 2026-10-26: 60 score · Scoring 3.0.0
  • 2026-10-27: 60 score · Scoring 3.0.0
  • 2026-10-28: 60 score · Scoring 3.0.0
  • 2026-10-29: 60 score · Scoring 3.0.0
  • 2026-10-30: 60 score · Scoring 3.0.0
  • 2026-10-31: 60 score · Scoring 3.0.0
  • 2026-11-01: 60 score · Scoring 3.0.0
  • 2026-11-02: 60 score · Scoring 3.0.0
  • 2026-11-03: 60 score · Scoring 3.0.0
  • 2026-11-04: 60 score · Scoring 3.0.0
  • 2026-11-05: 60 score · Scoring 3.0.0
  • 2026-11-06: 60 score · Scoring 3.0.0
  • 2026-11-07: 60 score · Scoring 3.0.0

multimodal category score history

2026-10-012026-11-07
  • 2026-10-01: 91 score · Scoring 3.0.0
  • 2026-10-02: 91 score · Scoring 3.0.0
  • 2026-10-03: 91 score · Scoring 3.0.0
  • 2026-10-04: 91 score · Scoring 3.0.0
  • 2026-10-05: 91 score · Scoring 3.0.0
  • 2026-10-06: 91 score · Scoring 3.0.0
  • 2026-10-07: 91 score · Scoring 3.0.0
  • 2026-10-08: 91 score · Scoring 3.0.0
  • 2026-10-09: 91 score · Scoring 3.0.0
  • 2026-10-10: 91 score · Scoring 3.0.0
  • 2026-10-11: 91 score · Scoring 3.0.0
  • 2026-10-12: 91 score · Scoring 3.0.0
  • 2026-10-13: 91 score · Scoring 3.0.0
  • 2026-10-14: 91 score · Scoring 3.0.0
  • 2026-10-15: 91 score · Scoring 3.0.0
  • 2026-10-16: 91 score · Scoring 3.0.0
  • 2026-10-17: 91 score · Scoring 3.0.0
  • 2026-10-18: 91 score · Scoring 3.0.0
  • 2026-10-19: 91 score · Scoring 3.0.0
  • 2026-10-20: 91 score · Scoring 3.0.0
  • 2026-10-21: 91 score · Scoring 3.0.0
  • 2026-10-22: 91 score · Scoring 3.0.0
  • 2026-10-23: 91 score · Scoring 3.0.0
  • 2026-10-24: 91 score · Scoring 3.0.0
  • 2026-10-25: 91 score · Scoring 3.0.0
  • 2026-10-26: 91 score · Scoring 3.0.0
  • 2026-10-27: 91 score · Scoring 3.0.0
  • 2026-10-28: 91 score · Scoring 3.0.0
  • 2026-10-29: 91 score · Scoring 3.0.0
  • 2026-10-30: 91 score · Scoring 3.0.0
  • 2026-10-31: 91 score · Scoring 3.0.0
  • 2026-11-01: 91 score · Scoring 3.0.0
  • 2026-11-02: 91 score · Scoring 3.0.0
  • 2026-11-03: 91 score · Scoring 3.0.0
  • 2026-11-04: 91 score · Scoring 3.0.0
  • 2026-11-05: 91 score · Scoring 3.0.0
  • 2026-11-06: 91 score · Scoring 3.0.0
  • 2026-11-07: 91 score · Scoring 3.0.0

07 / EXPLORE

Similar models