Tabbit
ResourcesBlogModels
Tabbit LogoTabbit

Tabbit — The AI Browser that Works for You

Topics

  • AI Browser Resources
  • Agentic Browser Resources
  • Browser Downloads and Install Guides
  • Browser Comparisons
  • AI Browser Alternatives
  • Browser Productivity Resources

Popular Guides

  • AI Browser
  • Agentic Browser Download
  • Best AI Browser 2026: Top 9 Tested & Ranked
  • AI Browser Download
  • Free AI Browser
  • Best AI Browser 2026
  • AI Browser Comparison 2026
  • AI Browser for Windows
  • AI Browser for Mac
  • Chrome Alternative 2026

Events

  • Tabbit Skill Competition
  • KPOP SBTI Fandom Personality Test
  • Tabbit Campus Creator Program
  • fifi's Picks: AI Skills for Research Papers
  • User Survey

About

  • Tabbit Blog
  • Press & Media
Review
MediaGemini 3.7 Flash

Gemini 3.7 Flash Benchmarks, Pricing & Speed

Original source

BenchLM.ai

AuthorBenchLM

Tabbit curation2026-08-19

Read original

Summary

Third-party benchmark aggregation page: a total score of 61.4/100, a public rank of #50/218, and coding, Agent, knowledge, mathematics, and pricing data.

Original article

The body text extracted during this visit is retained below; visible page text such as navigation, the platform's automatic translation, and comments is preserved as-is.


Skip to main content Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.

Start free brief BenchLM Leaderboard Radar Compare Benchmarks Models Research Newsletter Tools Search ⌘ K / Browse Models Gemini 3.7 Flash Gemini 3.7 Flash Current Released Aug 13, 2026 Proprietary Reasoning 1M context Google DeepMind Gemini 3.7 Flash model card

Released Aug 13, 2026 — see all recent releases

DECISION READING Gemini 3.7 Flash scores 61.4 out of 100 and ranks #50 of 218. This profile shows 15 source-displayable benchmark rows; its strongest eligible category is Coding at #12. API pricing is $0.75 input and $3.75 output per million tokens.

Data as of August 15, 2026 · How the score is built

BENCHLM SCORE

61.4

0 50 100

Public #50 of 218

Position in the ranked field

218 ranked median 52.7 Compare Gemini 3.7 Flash Find alternatives Strongest published evidence

Coding ranks #12. Particularly well-suited for software development and code generation tasks.

Validate before choosing

15 published rows leave some tracked benchmark slots empty. Knowledge is its lowest eligible category at #32.

Read this first. The public rank is available, but this profile does not yet have enough sourced coverage for a verified position.

On this page Decision Capability Cost Evidence Specs Ledger Lineage Notes FAQ

RADAR

Track Gemini 3.7 Flash's lifecycle.

Watch Google for a published repricing, replacement, or deprecation that affects this model.

Compare Google Watch Decision snapshot

Each value carries a field reference instead of floating alone. Markers compare this model with the current ranked and priced catalog; they are not absolute quality thresholds.

CAPABILITY

61.4 /100

field median 58.2

#50 of 218 ranked models

PRICE

$0.75 input / $3.75 output

input median $1

blended $2.25

SPEED

340 tok/s

field median 94 tok/s

First token 9.83 s

CONTEXT

1M tokens

field median 200,000

Maximum output length is tracked separately

Capability shape

Each axis shows percentile within that category’s eligible cohort. The comparison outline is the median of the six nearest public-score peers; a collapsed vertex means the category is not rank-eligible.

AGENTIC CODING REASONING KNOWLEDGE MATH MULTILINGUAL MULTIMODAL INSTRUCTION FOLLOWING Eligible category ranks Agentic #25/130 Coding #12/135 Reasoning Not ranked Knowledge #32/57 Math Not ranked Multilingual Not ranked Multimodal Not ranked Inst. Following Not ranked Top decile Top quartile Mid-field Not eligible What it costs to get this score

Published API price against the public score. The x-axis uses a log scale; the dashed path marks models that are not beaten by a cheaper, higher-scoring option. Price uses average of published input and output rates.

Explore all models Current model Gemini 3.7 Flash · 61.4 score · $2.25 blended per million tokens 40 50 60 70 80 90 $0.50 $1 $5 $10 $25 ↘ frontier Gemini 3.7 Flash

Horizontal: blended price per million tokens, log scale · Vertical: public score

How much of this is verified

Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.

Agentic 5/5 verified Coding 3/3 verified Reasoning 1/1 verified Knowledge 2/2 verified Math Not measured Multilingual Not measured Multimodal 4/4 verified Inst. Following Not measured Verified source Provisional Not measured Spec sheet

Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.

API model ID gemini-3.7-flash Google Gemini 3.7 Flash model documentation Context window 1M Google Gemini 3.7 Flash model documentation Maximum output Not sourced yet Knowledge cutoff Not sourced yet Input modalities text, image, video, audio, pdf Google Gemini 3.7 Flash model documentation Output modalities text Google Gemini 3.7 Flash model documentation Parameters Not disclosed by the provider Availability Gemini API · Google AI Studio · Google Antigravity · Gemini Enterprise Agent Platform · Gemini Enterprise · Gemini app via Spark Google Gemini 3.7 Flash launch Cloud regions Not tracked yet Lifecycle active Google Gemini 3.7 Flash model documentation API capabilities Tool calling, structured outputs, and batch support are not tracked yet Prompt caching Not documented in the pricing record Google: Introducing Gemini 3.7 Flash Self-host Weights are not published Rate limits Not tracked yet Category score record

Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.

Category scores, ranks, weighting, benchmark coverage, and evidence status Category Score Rank Percentile Weight Benchmarks Evidence

Agentic 58.0 #25 of 130 81st 22% 5 benchmarks Verified

Coding 66.4 #12 of 135 92nd 20% 3 benchmarks Verified

Reasoning Score pending Not ranked Not available 17% 1 benchmark Verified

Knowledge 67.4 #32 of 57 45th 12% 2 benchmarks Verified

Math Not measured Not ranked Not available 5% 0 benchmarks Not measured

Multilingual Not measured Not ranked Not available 7% 0 benchmarks Not measured

Multimodal 76.9 Not ranked Not available 12% 4 benchmarks Verified

Inst. Following Not measured Not ranked Not available 5% 0 benchmarks Not measured Benchmark ledger

Coding opens by default. The marker compares each value with the best source-verified result in the catalog; provisional leaders do not set the reference. Expand the remaining categories for every published row.

Coding 3 rows Coding benchmark values, best verified comparison, weight, and source status Benchmark Score Versus best verified row Gap Weight Evidence

FrontierCode 1.1 Main 43.6%

Best verified: Claude Fable 5 · 53.5%

9.9 behind	Display only	

Provider exact Google DeepMind: Gemini 3.7 Flash model card

deepSwe 65.3%

Best verified: GPT-5.6 Sol · 72.7%

7.4 behind	Display only	

Provider exact Google DeepMind: Gemini 3.7 Flash model card

Terminal-Bench 2.1 Terminal-Bench 2.1 (provider run) 85.8%

Best verified: GLM-5.3 · 88.2%

2.4 behind	Display only	

Provider exact Google DeepMind: Gemini 3.7 Flash model card Agentic 5 rows Reasoning 1 row Knowledge 2 rows Multimodal 4 rows Lineage

The sequence follows explicit supersedes links. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.

December 2025

Gemini 3 Flash

Score 59.7 · $0.5 / $3

May 19, 2026

Gemini 3.5 Flash

Score 64.7 · $1.5 / $9

Jul 21, 2026

Gemini 3.6 Flash

Score 75.5 · $1.5 / $7.5

Aug 13, 2026 · you are here

Gemini 3.7 Flash

Score 61.4 · $0.75 / $3.75

Base entry

How to read this profile

The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.

Gemini 3.7 Flash ranks #50 of 218 on the public leaderboard with a score of 61.36/100. It does not yet have enough sourced coverage for a verified position.

Gemini 3.7 Flash is a proprietary model with a 1M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.

Available as the stable gemini-3.7-flash model through the Gemini API and Google AI Studio, and through Google Antigravity, Gemini Enterprise Agent Platform, Gemini Enterprise, and Gemini Spark. Google documents text, image, video, audio, and PDF input; text output; a 1,048,576-token input limit; and a 65,536-token output limit.

Its explicit predecessor is Gemini 3.6 Flash. 15 of 437 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.

Its strongest eligible category is Coding at #12, while its lowest eligible position is Knowledge at #32. particularly well-suited for software development and code generation tasks.

RADAR Gemini 3.7 Flash release history Full release history Gemini 3.7 Flash released

Google · Model release

Source confirmed Aug 13, 2026 Frequently asked questions How does Gemini 3.7 Flash perform overall in AI benchmarks? + Is Gemini 3.7 Flash good for knowledge and understanding? + Is Gemini 3.7 Flash good for coding and programming? + Is Gemini 3.7 Flash good for reasoning and logic? + Is Gemini 3.7 Flash good for agentic tool use and computer tasks? + Is Gemini 3.7 Flash good for multimodal and grounded tasks? + Does Gemini 3.7 Flash have full benchmark coverage on BenchLM? + What is the context window size of Gemini 3.7 Flash? +

Related resources

More from Google Provider catalog, models, and evidence. LLM pricing comparison Compare published API rates across providers. Benchmark methodology How scores, ranks, and evidence states are built. Which LLM should I use? Filter the catalog by task and operating constraints.

Last updated August 15, 2026. Runtime fields remain blank until a sourced snapshot exists.

Watch Gemini 3.7 Flash in the weekly brief

Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.

Read a sample issue

Join 2,000+ readers.

Leave this field blank Send me the weekly brief

By clicking “Send me the weekly brief,” you agree to receive one BenchLM email each week. Unsubscribe anytime. Privacy

BenchLM

Independent LLM benchmarks, pricing, and runtime evidence. Published results expose their evidence path and review date.

✓ Supported and Estimated evidence labels

Models tracked 394 Benchmarks 437 Data refreshed August 15, 2026

Tracked means a model profile exists. Supported positions have sufficiently diverse direct evidence. Estimated positions remain ranked with wider uncertainty while evidence is incomplete.

How BenchLM ranks models Explore LLM leaderboard Compare models Models Benchmarks Providers Best models by use case Model alternatives Rankings Frontier models Coding models Agentic models Reasoning models Multimodal models Best overall Best open source Best Chinese models Dashboards Free Radar Brief Model updates Release timeline AI Race LLM pricing Price vs performance LLM speed Agent benchmarks Pricing trends LLM statistics Research & tools Research Weekly brief All tools AI prompt writer Prompt optimizer LLM selector Token counter Download data Methodology Evidence confidence Leaderboard history AI app stack

Raw benchmark rows stay exact; composite rankings use disclosed missing-data priors. © 2026 BenchLM.

About Advertise Contact Editorial policy Affiliate disclosure Privacy Terms llms.txt RSS


Usage notes

This document is an archive of source material and does not represent an endorsement of the original article's conclusions by Tabbit or the maintainer of this document. When citing benchmark scores, prices, or model capabilities, return to the original source to confirm the version, test set, and date.

Curated by Tabbit

This is a third-party source navigator. Model versions, test environments, and personal experience vary; consult the original source.

Gemini 3.7 Flash

Use and compare models in Tabbit

Gemini 3.7 Flash

Related reviews

OfficialGoogle DeepMind2026-08-13

Google DeepMind Gemini 3.7 Flash Model Card

CommunityReddit r/GeminiAI2026-08-16

3.7 Flash feels insanely fast — but is it hallucinating more than 3.6?

Mediaeesel AI Blog2026-08-14

Gemini 3.7 Flash: Real Pricing, Speed, and Agent Boundaries (eesel)

MediaArs Technica2026-08-14

Google announces Gemini 3.7 Flash

Gemini 3.7 Flash

Related prompts

OfficialGoogle Blog

Introducing Gemini 3.7 Flash — coding and agent workflows

CommunityX

Gemini 3.7 Flash image prompt — causal visual description

MediaTom’s Guide2026-08-14

Gemini 3.7 Flash is here — and this prompt proves why it is Google’s workhorse

CommunityX

Gemini 3.7 Flash with one prompt — voxel Japanese pagoda garden