VerdictPal
Join

Models / Google

Gemini 3 Flash Preview (Reasoning)

Gemini 3 · Released 2025-12-17

AA Intelligence Index 37.8. Listed on Artificial Analysis with API pricing $0.5/$3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#71 of 184 on AA Index · current snapshot
26 AA Index · 12 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite63
Context128K
Input / 1M$0.50
Output / 1M$3.0
Knowledge cutoffopen
Statussolid

Family profile

Best published score in each covered benchmark family.

81/100 avg
Knowledge
89
Reasoning
90
Math
97
Coding
91
Agentic
39
Long context
78

6 tested benchmark families

Benchmark placements

Where Gemini 3 Flash Preview (Reasoning) places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
AIMEMath#2/ 3797Artificial Analysis2026-09-10
LiveCodeBenchCoding#2/ 3191Artificial Analysis2026-09-10
MMLU-ProKnowledge#4/ 3989Artificial Analysis2026-09-10
IFBenchReasoning#11/ 12778Artificial Analysis2026-09-10
GPQA DiamondReasoning#48/ 19090Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#52/ 18537Artificial Analysis2026-09-10
AA-LCRLong context#60/ 18278Artificial Analysis2026-09-10
τ³-Bench BankingAgentic#61/ 11121Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#77/ 18826Artificial Analysis2026-09-10
AA output speedPerformance#112/ 190Artificial Analysis2026-09-10
AA time to first tokenPerformance#112/ 190Artificial Analysis2026-09-10
Terminal-BenchAgentic#124/ 18939Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Gemini 3 Flash Preview (Reasoning) shows up, it's highlighted.

KnowledgeGemini 3 Flash Preview (Reasoning) · 89
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningGemini 3 Flash Preview (Reasoning) · 90
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
MathGemini 3 Flash Preview (Reasoning) · 97
  1. 1
    OpenAI o3
    99
  2. 2
    GPT-5 (medium)
    99
  3. 3
    Grok 4
    99
CodingGemini 3 Flash Preview (Reasoning) · 91
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGemini 3 Flash Preview (Reasoning) · 39
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGemini 3 Flash Preview (Reasoning) · 78
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

Benchmark rows added to the public ledger for Gemini 3 Flash Preview (Reasoning) in the last 120 days. Older rows live in the full table below.

12 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context78%Artificial Analysis4
AA output speedPerformance0 t/sArtificial Analysis4
AA time to first tokenPerformance0.00sArtificial Analysis4
AIMEMath97%Artificial Analysis4
Artificial Analysis Intelligence IndexReasoning26.3%Artificial Analysis4
GPQA DiamondReasoning89.8%Artificial Analysis4
Humanity's Last ExamReasoning36.6%Artificial Analysis4
IFBenchReasoning77.96%Artificial Analysis4
LiveCodeBenchCoding90.8%Artificial Analysis4
MMLU-ProKnowledge89%Artificial Analysis4
τ³-Bench BankingAgentic20.82%Artificial Analysis4
Terminal-BenchAgentic38.64%Artificial Analysis4
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge89
Reasoning90
Math97
Coding91
Agentic39
Long context78
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Gemini 3 Flash Preview (Reasoning). Scores link to the original source; gaps mean no public row exists yet.

12 sourced rows

12 of 37 catalog benchmarks have a sourced row for Gemini 3 Flash Preview (Reasoning).

Knowledge

Reasoning

Math

Coding

Agentic

Long context

Performance

Sources