Skip to content
← All models
Google

Gemini 3.7 Flash benchmarks & scores

Google Proprietary Released Aug 2026 Official website
62.5
Vibe Coding Index · #9 of 103

Independent Gemini 3.7 Flash AI coding benchmark results Intelligence 56, Coding 79, Agentic 45, with a Vibe Coding Index of 62.5. Compare its overall rank, token price, context window, and practical agentic workflow strengths below.

Blended price #65/120
$0.76
$/M tokens · 3 to 1 in, out
Context
1M
Window size

Evidence confidence

Medium confidence

100% core coverage

Complete core scores supported by one current verified source.

3/3
Core scores
2
Evidence sources
Aug 14
Last measured
Arena code webdev → Artificial analysis →

Catalog capabilities

What the model supports

1M context
Include reasoning Max tokens Reasoning Reasoning effort Response format Seed Stop Structured outputs Temperature Tool choice
Input
Text, Image, Video, File, Audio
Output
Text
Knowledge cutoff
Not listed
Cache read price
$0.04 per million tokens

Suite radar

Intelligence Coding Agentic

Suite scores

Intelligence #11/103
General reasoning & knowledge (Intelligence Index)
56.0
Coding #6/134
Code generation & software tasks (Coding Index)
78.9
Agentic #18/104
Multi-step tool use & agentic workflows (Agentic Index)
45.1

Task placements

Where Gemini 3.7 Flash places outside the VCI blend

Head-to-head preference boards that never rewrite the suite scores above — Arena Code WebDev and Design Arena categories only.

WebDev board → Design Arena →

Arena Code WebDev

Preliminary

Human preference for web development outputs. Separate from Artificial Analysis coding until blended into the published Coding suite score.

Coding suite score above (78.9, multi-source index) includes this signal when current.

Rank
#6 /54
ELO
1,588 ±13
Source #
8
Votes
2,544

Design Arena categories

Task leaderboards for Gemini 3.7 Flash

Website, UI, game, data viz, and other tournament boards — not used in the Vibe Coding Index formula.

No Design Arena category row is matched to Gemini 3.7 Flash yet. Browse the web development and other category boards for the wider field.

Source record

Source. Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings).

Artificial Analysis via OpenRouter OpenRouter model card Arena Code WebDev

TheVibeFather rankings

Gemini 3.7 Flash practical agentic coding scorecard

This is our algorithmic assessment of where Gemini 3.7 Flash ranks for common vibe coding and agentic engineering workflows. Every stable 0–10 score is calculated consistently from the sourced Intelligence, Coding, and Agentic signals above—never from an unsourced opinion score.

Overall VibeFather rating
6.3 /10
#9 of 103 models
Practical area

Small, well-defined code changes

Precision on scoped edits, fixes, and implementation tasks.

Strong 7.2/10
Field rank #6 / 103

UI and CSS iteration

Front-end implementation with iterative tool-driven refinement.

Strong 7.2/10
Field rank #7 / 104

Routine debugging

Diagnosing failures and turning reasoning into correct code changes.

Strong 6.6/10
Field rank #7 / 103

Repository-wide refactors

Coordinating larger edits across files while preserving intent.

Strong 6.3/10
Field rank #9 / 103

Architecture decisions

Reasoning through tradeoffs, constraints, and system-level choices.

Capable 5.2/10
Field rank #13 / 103

Tool use and workflow execution

Planning and completing multi-step work with tools and feedback loops.

Capable 5.4/10
Field rank #16 / 104

Long autonomous coding tasks

Sustaining coherent progress across longer agentic engineering runs.

Capable 4.7/10
Field rank #17 / 103

Overall VibeFather rating

Our complete Vibe Coding Index, expressed on the same 0–10 practical scale.

Strong 6.3/10
Field rank #9 / 103
1 · Sourced signals

We start with Intelligence, Coding, Agentic benchmark evidence, and preserve missing values as unverified.

2 · Practical lenses

Our code-controlled algorithm blends the native 0–100 signals by workflow and expresses the result on a stable 0–10 scale. Every required input must be present.

3 · Field ranking

Each workflow score is ranked against every model with comparable evidence, so positions update when the benchmark field changes.

The Vibe Coding Index is TheVibeFather’s code-controlled 0–100 composite for practical coding and agentic capability. Price and adoption are reported independently and never alter quality. Exact weighting remains proprietary, inputs, missing-data behavior, and per-category ranks are disclosed here. Read the benchmark methodology →

Recorded score history

How Gemini 3.7 Flash has moved

All benchmark history →
65.0 62.5 60.0
Aug 13, 2026 9 PM CT
62.5 rank 9

No material score or rank change.

Aug 13, 2026 8 PM CT
62.5 rank 9

Baseline snapshot recorded.

Where Gemini 3.7 Flash sits

Its bar stays full-color on each index while the rest of the field recedes. The price scatter below keeps the same highlight.

Intelligence

Artificial Analysis Intelligence Index · Higher is better

63.1 Claude Opus 5 Claude Opus 5 Anthropic #1 of 103 · Intelligence 63.1 Intelligence 63.1 Coding 85.2 Agentic 59.2 VCI 71.7 62.1 Claude Fable 5 Claude Fable 5 Anthropic #2 of 103 · Intelligence 62.1 Intelligence 62.1 Coding 81.1 Agentic 56.6 VCI 68.7 60.9 Grok 4.6 Grok 4.6 xAI #3 of 103 · Intelligence 60.9 Intelligence 60.9 Coding 76.8 Agentic 58.7 VCI 67.3 59.7 Kimi K3 Kimi K3 Moonshot #4 of 103 · Intelligence 59.7 Intelligence 59.7 Coding 83.2 Agentic 54.3 VCI 68.4 59 GPT-5.6 Sol GPT-5.6 Sol OpenAI #5 of 103 · Intelligence 59 Intelligence 59 Coding 82 Agentic 53.6 VCI 67.5 58.1 Qwen3.8 Max Qwen3.8 Max Alibaba #6 of 103 · Intelligence 58.1 Intelligence 58.1 Coding 80.1 Agentic 58.4 VCI 68.1 57.3 Claude Opus 4.8 Claude Opus 4.8 Anthropic #7 of 103 · Intelligence 57.3 Intelligence 57.3 Coding 76.5 Agentic 49.4 VCI 63.2 57 Muse Spark 1.2 Muse Spark 1.2 Meta #8 of 103 · Intelligence 57 Intelligence 57 Coding 71.8 Agentic 48.4 VCI 60.7 56.6 GPT-5.6 Terra GPT-5.6 Terra OpenAI #9 of 103 · Intelligence 56.6 Intelligence 56.6 Coding 76 Agentic 50.2 VCI 63.1 56.3 GPT-5.5 GPT-5.5 OpenAI #10 of 103 · Intelligence 56.3 Intelligence 56.3 Coding 74.1 Agentic 47.4 VCI 61.2 56 Gemini 3.7 Flash Gemini 3.7 Flash Google #11 of 103 · Intelligence 56 Intelligence 56 Coding 78.9 Agentic 45.1 VCI 62.5 55.8 Grok 4.5 Grok 4.5 xAI #12 of 103 · Intelligence 55.8 Intelligence 55.8 Coding 74.7 Agentic 48.9 VCI 61.9 55.3 Claude Sonnet 5 Claude Sonnet 5 Anthropic #13 of 103 · Intelligence 55.3 Intelligence 55.3 Coding 73.5 Agentic 49.7 VCI 61.5 55 Claude Opus 4.7 Claude Opus 4.7 Anthropic #14 of 103 · Intelligence 55 Intelligence 55 Coding 75.7 Agentic 46.3 VCI 61.3 53.2 DeepSeek V4 Pro 0813 DeepSeek V4 Pro 0813 DeepSeek #15 of 103 · Intelligence 53.2 Intelligence 53.2 Coding 68.8 Agentic 49.6 VCI 59 53.2 Muse Spark 1.1 Muse Spark 1.1 Meta #16 of 103 · Intelligence 53.2 Intelligence 53.2 Coding 73.3 Agentic 39.7 VCI 57.5 53.1 GPT-5.4 GPT-5.4 OpenAI #17 of 103 · Intelligence 53.1 Intelligence 53.1 Coding 69.4 Agentic 44.2 VCI 57.3 52.6 GLM-5.2 GLM-5.2 Zhipu #18 of 103 · Intelligence 52.6 Intelligence 52.6 Coding 74.1 Agentic 45.7 VCI 59.9 52.3 GPT-5.6 Luna GPT-5.6 Luna OpenAI #19 of 103 · Intelligence 52.3 Intelligence 52.3 Coding 72.3 Agentic 46.9 VCI 59.4 52 Gemini 3.5 Flash Gemini 3.5 Flash Google #20 of 103 · Intelligence 52 Intelligence 52 Coding 70.9 Agentic 39.7 VCI 56.2 51.8 DeepSeek V4 Flash 0731 DeepSeek V4 Flash 0731 DeepSeek #21 of 103 · Intelligence 51.8 Intelligence 51.8 Coding 69.1 Agentic 48.4 VCI 58.4 51.6 Gemini 3.6 Flash Gemini 3.6 Flash Google #22 of 103 · Intelligence 51.6 Intelligence 51.6 Coding 71.9 Agentic 40.5 VCI 56.9 48.4 Claude Sonnet 4.6 Claude Sonnet 4.6 Anthropic #23 of 103 · Intelligence 48.4 Intelligence 48.4 Coding 67.2 Agentic 42.1 VCI 54.7 47.7 Gemini 3.1 Pro Preview Gemini 3.1 Pro Preview Google #24 of 103 · Intelligence 47.7 Intelligence 47.7 Coding 67.1 Agentic 23 VCI 47.8 46.7 Qwen3.7 Max Qwen3.7 Max Alibaba #25 of 103 · Intelligence 46.7 Intelligence 46.7 Coding 68.8 Agentic 30.9 VCI 51.1 45.4 MiniMax-M3 MiniMax-M3 MiniMax #26 of 103 · Intelligence 45.4 Intelligence 45.4 Coding 62.6 Agentic 36.1 VCI 49.9 45.1 Kimi K2.6 Kimi K2.6 Moonshot #27 of 103 · Intelligence 45.1 Intelligence 45.1 Coding 65.6 Agentic 31.2 VCI 49.5 43 Kimi K2.7 Code Kimi K2.7 Code Moonshot #28 of 103 · Intelligence 43 Intelligence 43 Coding 63.2 Agentic 30.3 VCI 47.6
Reasoning models are marked with a lightbulb Full board →
↖ Best value Premium ↗ ↙ Budget Poor value ↘ 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 $0.01 $0.03 $0.1 $0.32 $1 $3.16 $10 $31.62 BLENDED COST — $ / 1M TOKENS (LOG) VIBE CODING INDEX #1 Claude Opus 5 #3 Kimi K3 Qwen3.8 Max GPT-5.6 Terra Gemini 3.7 Flash GPT-5.6 Luna Hy3 Solar Pro 4 #69 North Mini Code Nemotron 3.5 Lightning
Efficient frontier Dashed guides = field medians 40 model(s) hidden — data not yet verified
Vendor key (21)
Alibaba Anthropic Cohere DeepSeek Google Inception Inclusionai Meta Meta Llama MiniMax Mistral Moonshot NVIDIA OpenAI Stepfun Tencent Thinkingmachines Upstage Xiaomi Zhipu xAI
Google DeepMind

Company pulse

Latest from Google DeepMind

Google DeepMind Company ↗ Demis Hassabis CEO ↗

Up to ten recent public posts from the company and its leadership, linked to the originals on X.

Loading recent posts…

Public posts are cached briefly for speed.

Model benchmark FAQ

Gemini 3.7 Flash for vibe coding and agentic engineering

Answers update from this model’s current sourced scores, practical rankings, and verified pricing.

What is Gemini 3.7 Flash's Vibe Coding Index?

Gemini 3.7 Flash has a Vibe Coding Index of 62.5/100 and is #9 of 103 in the current field. The index is TheVibeFather's stable composite for practical AI coding, combining independently sourced Intelligence, Coding, and Agentic benchmark signals. Price and adoption are compared separately and never inflate the quality score.

How does Gemini 3.7 Flash rank for vibe coding?

For vibe coding, Gemini 3.7 Flash's strongest practical area is Small, well-defined code changes at 7.2/10, ranking #6 of 103 models in TheVibeFather rankings.

Is Gemini 3.7 Flash good for agentic coding and autonomous tasks?

Gemini 3.7 Flash scores 5.4/10 for tool use and workflow execution and 4.7/10 for long autonomous coding tasks. Those categories emphasize the Agentic benchmark signal used for multi-step planning, tool calls, edits, tests, and feedback loops.

What coding benchmark scores does Gemini 3.7 Flash have?

The current sourced benchmark profile for Gemini 3.7 Flash is Intelligence 56/100, Coding 79/100, Agentic 45/100. TheVibeFather converts those 0–100 signals into consistent practical 0–10 workflow scores so models can be compared for real agentic engineering work.

How much does Gemini 3.7 Flash cost for agentic engineering?

The current benchmark uses a 3 to 1 input-to-output blend of $0.76 per million blended tokens. That places it #65 of 120 by price, where a lower rank number means less expensive.

Are vibe coding, agentic coding, and agentic engineering the same?

They describe overlapping AI-assisted software workflows. Vibe coding emphasizes directing software through intent and iteration, agentic coding emphasizes a model planning and executing multi-step work with tools, agentic engineering is the broader discipline of designing, supervising, and validating those workflows. TheVibeFather benchmarks the shared capabilities behind all three terms on every model page, including Gemini 3.7 Flash.

Compare Gemini 3.7 Flash → Back to leaderboard