VerdictPal
Join

Models / Google

Gemini 3.7 Flash

Gemini 3.7 · Released 2026-08-13

Google's August 2026 Flash workhorse. Artificial Analysis Intelligence Index 56 on the high-effort endpoint. Introductory API price $0.75 / $3.75 per 1M tokens through 31 Dec 2026.

Current Google Flash model for coding agents, knowledge-work runs, and high-throughput tool use. Prefer sourced ledger rows over launch claims; AA scores sync from the public snapshot.

#11 of 166 on AA Index · current snapshot
56 AA Index · 10 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite67
Context1M
Input / 1M$0.75
Output / 1M$3.8
Knowledge cutoffunpublished
Output speed307 t/s
TTFT (default API)7.00s
Statusflagship
Gemini

Family profile

Best published score in each covered benchmark family.

81/100 avg
Reasoning
95
Coding
57
Agentic
86
Multimodal
86
Long context
80

5 tested benchmark families

Editor's note
Recorded from the 2026-08-13 AA snapshot (slug gemini-3-7-flash, high effort, index 56) plus the Google launch post and DeepMind model card (1M context, 64K output). Editorial depth and failure modes still need a desk pass — scores below are ledger-backed, not hands-on claims. Google-reported FrontierCode, DeepSWE, WebDev Arena, and GDP.pdf figures stay vendor-claimed.
Benchmark placements

Where Gemini 3.7 Flash 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
MMMU-ProMultimodal#1/ 886Artificial Analysis — MMMU-Pro2026-09-01
GPQA DiamondReasoning#3/ 18595Artificial Analysis2026-09-01
SciCodeCoding#5/ 18057Artificial Analysis2026-09-01
AA-LCRLong context#6/ 17980Artificial Analysis2026-09-01
AA output speedPerformance#6/ 185Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#8/ 18048Artificial Analysis2026-09-01
Terminal-BenchAgentic#9/ 18486Artificial Analysis2026-09-01
AA time to first tokenPerformance#16/ 185Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#18/ 18456Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#32/ 10633Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Gemini 3.7 Flash shows up, it's highlighted.

ReasoningGemini 3.7 Flash · 95
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingGemini 3.7 Flash · 57
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticGemini 3.7 Flash · 86
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextGemini 3.7 Flash · 80
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Gemini 3.7 Flash in the last 120 days. Older rows live in the full table below.

10 recent
BenchmarkFamilyScoreSourceDays ago
MMMU-ProMultimodal85.5%Artificial Analysis — MMMU-Pro2
AA-LCRLong context80%Artificial Analysis2
AA output speedPerformance307 t/sArtificial Analysis2
AA time to first tokenPerformance7.00sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning56%Artificial Analysis2
GPQA DiamondReasoning94.5%Artificial Analysis2
Humanity's Last ExamReasoning47.9%Artificial Analysis2
SciCodeCoding56.8%Artificial Analysis2
τ³-Bench BankingAgentic32.78%Artificial Analysis2
Terminal-BenchAgentic85.77%Artificial Analysis2
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.

Reasoning95
Coding57
Agentic86
Multimodal86
Long context80
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Gemini 3.7 Flash. Scores link to the original source; gaps mean no public row exists yet.

10 sourced rows

10 of 37 catalog benchmarks have a sourced row for Gemini 3.7 Flash.

Reasoning

Coding

Agentic

Multimodal

Long context

Performance

Changelog
  1. Gemini 3.7 Flash generally available. AA high-effort Intelligence Index 56; intro price $0.75 / $3.75 per 1M tokens. Google reports gains over 3.6 Flash on FrontierCode 1.1, DeepSWE v1.1, WebDev Arena, and GDP.pdf — vendor-claimed, not ledger rows.
Sources