Gemini 3.7 Flash released
Google · Model release
Source confirmed
Data as of September 27, 2026 · How the score is built
Superseded.Google has newer models in this line:Gemini 3.8 Flash
SupersededProprietaryReasoning
Released Aug 13, 20261M contextGoogle DeepMind Gemini 3.7 Flash model card
Gemini 3.7 Flash
Decision readingGemini 3.7 Flash scores 67.7 out of 100 and ranks #19 of 194. This profile shows 25 source-displayable benchmark rows; its strongest eligible category is Multimodal & Grounded at #10. API pricing is $0.75 input and $3.75 output per million tokens, with cached input at $0.075.
Released Aug 13, 2026 — see all recent releases
Gemini 3.7 Flash will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Share or export
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
67.7/100
field median 50.2#19 of 194 ranked models
Public
#19of 194
Verified #14 of 74
Price
$0.75input / $3.75 output
input median $0.97cached $0.075 · blended $2.25
Speed
278tok/s
field median 91 tok/sFirst token 11.4 s
Context
1Mtokens
field median 256,000Maximum output length is tracked separately
Strongest published evidence
Multimodal & Grounded ranks #10. Particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Validate before choosing
25 published rows leave some tracked benchmark slots empty. Agentic is its lowest eligible category at #20.
Source-linked · 25 displayable benchmark rows
Follow model changesCategory 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 | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #20 of 105Percentile 82ndWeight 22%7 benchmarksVerified | 58.6 | 7 benchmarks | Verified | |||
| CodingRank #17 of 135Percentile 88thWeight 20%6 benchmarksVerified | 60.3 | 6 benchmarks | Verified | |||
| ReasoningRank Not rankedWeight 17%3 benchmarksVerified | 77.8 | 3 benchmarks | Verified | |||
| MultimodalRank #10 of 50Percentile 82ndWeight 12%3 benchmarksVerified | 82.6 | 3 benchmarks | Verified | |||
| KnowledgeRank #10 of 158Percentile 94thWeight 12%6 benchmarksVerified | 71.1 | 6 benchmarks | Verified | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
25 of 486 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow 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.
- Agentic7/7 verified
- Coding6/6 verified
- Reasoning3/3 verified
- Multimodal3/3 verified
- Knowledge6/6 verified
- MultilingualNot measured
- Inst. FollowingNot measured
- MathNot measured
Verified sourceProvisionalNot measured
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.
Gemini 3.7 Flash category percentile values
- Agentic82nd percentile
- Coding88th percentile
- ReasoningNot eligible
- Multimodal82nd percentile
- Knowledge94th percentile
- MultilingualNot eligible
- Instruction followingNot eligible
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#20/105
- Coding#17/135
- ReasoningNot ranked
- Multimodal#10/50
- Knowledge#10/158
- MultilingualNot ranked
- Inst. FollowingNot ranked
- MathNot ranked
Top decileTop quartileMid-fieldNot eligible
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.
Coding6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| DeepSWE | Score65.3% | Versus best verified row Best verified: Muse Spark 1.3 · 75.4% | Gap10.1 behind | Weight15% ref. weight | Provider exact |
| FrontierSWE v2 | Score20.3% | Versus best verified row Best verified: GPT-6 Astra · 65.5% | Gap45.2 behind | Weight8% ref. weight | Benchmark exact |
| LiveCodeBench (Vals)LiveCodeBench, Vals AI run | Score88.7% | Versus best verified row Best verified: Claude Fable 5.1 · 90.5% | Gap1.8 behind | Weight8% ref. weight | |
| FrontierCode 1.1 Main | Score43.6% | Versus best verified row Best verified: Claude Opus 5.5 · 54.4% | Gap10.8 behind | Weight4% ref. weight | Provider exact |
| Terminal-Bench 2.1 | Score85.8% | Versus best verified row Best verified: SWE-2 · 92.8% | Gap7 behind | WeightScored in Agentic | Provider exact |
| SWE-bench (Vals)SWE-bench, Vals AI run | Score80.8% | Versus best verified row Best verified: Claude Opus 5 · 97.0% | Gap16.2 behind | WeightDisplay only | Verified |
Agentic7 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| OSWorld 2.0 | Score47.9% | Versus best verified row Best verified: GPT-6 Astra · 72.6% | Gap24.7 behind | Weight10% ref. weight | Provider exact |
| Terminal-Bench 2.1 | Score85.8% | Versus best verified row Best verified: SWE-2 · 92.8% | Gap7 behind | Weight8% ref. weight | Provider exact |
| AutomationBench | Score30.4% | Versus best verified row Best verified: DeepSeek V4.1 Flash · 54.8% | Gap24.4 behind | Weight5% ref. weight | Provider exact |
| Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run | Score77.5% | Versus best verified row Best verified: GPT-6 Astra · 87.3% | Gap9.8 behind | Weight3% ref. weight | |
| Terminal-Bench 3.0 | Score14.9% | Versus best verified row Best verified: Claude Opus 5 · 42.7% | Gap27.8 behind | Weight3% ref. weight | Provider exact |
| Agents' Last Exam | Score26.3% | Versus best verified row Best verified: GPT-6 Astra · 59.3% | Gap33 behind | Weight3% ref. weight | Provider exact |
| ApprenticeBenchApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job | Score16% | Versus best verified row Best verified: Claude Fable 5.1 · 72% | Gap56 behind | WeightDisplay only | Benchmark exact |
Reasoning3 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| ARC-AGI-2Abstraction and Reasoning Corpus for AGI v2 | Score84.6% | Versus best verified row Best verified: GPT-6 Astra · 95% | Gap10.4 behind | WeightWeighted 25% | Benchmark exact |
| MRCR v2 64K-128KOpenAI MRCR v2 8-needle 64K-128K | Score97% | Versus best verified row Best verified: Gemini 3.7 Flash · 97% | GapBest verified | WeightDisplay only | Provider exact |
| ARC-AGI-1ARC-AGI-1 Semi-Private Evaluation | Score95.50% | Versus best verified row Best verified: GPT-6 Astra · 98.50% | Gap3 behind | WeightDisplay only | Benchmark exact |
Multimodal3 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| CharXivCharXiv Reasoning | Score88.7% | Versus best verified row Best verified: Qwen3.8 Max · 93.5% | Gap4.8 behind | WeightWeighted 20% | Provider exact |
| CharXiv w/o toolsCharXiv Reasoning without tools | Score84.5% | Versus best verified row Best verified: Claude Mythos 5 · 88.9% | Gap4.4 behind | WeightDisplay only | Provider exact |
| LVBench | Score85.4% | Versus best verified row Best verified: Gemini 3.8 Flash · 87.1% | Gap1.7 behind | WeightDisplay only | Provider exact |
Knowledge6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMLU-Pro (Vals)MMLU-Pro, Vals AI run | Score90.1% | Versus best verified row Best verified: Claude Fable 5.1 · 92.4% | Gap2.3 behind | Weight6% ref. weight | Verified |
| GPQA Diamond (Vals)GPQA Diamond, Vals AI run | Score93.9% | Versus best verified row Best verified: Gemini 3.1 Pro · 95.5% | Gap1.6 behind | Weight2% ref. weight | |
| HLE-Verified | Score53.6% | Versus best verified row Best verified: Gemini 3.8 Flash · 54.9% | Gap1.3 behind | WeightDisplay only | Provider exact |
| LABBench2LABBench2: An Improved Benchmark for AI Systems Performing Biology Research | Score82.1% | Versus best verified row Best verified: Gemini 3.8 Flash · 86.2% | Gap4.1 behind | WeightDisplay only | Provider exact |
| BioMysteryBench (human-solvable)BioMysteryBench Human Solvable | Score87.1% | Versus best verified row Best verified: Claude Opus 5 · 90.1% | Gap3 behind | WeightDisplay only | Provider exact |
| BioMysteryBench (human-difficult)BioMysteryBench Human Difficult | Score43.5% | Versus best verified row Best verified: Gemini 3.8 Flash · 56.5% | Gap13 behind | WeightDisplay only | Provider exact |
Bars run 0–100; the dark tick marks the best source-verified value
All 25 rowsWhat 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.
Current modelExplore all models
Gemini 3.7 Flash · 67.7 score · $2.25 blended per million tokens
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Horizontal: blended price per million tokens, log scale · Vertical: public score
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Radar
Full release historyGemini 3.7 Flash release history
Radar confirmed these at the source. Use Gemini 3.7 Flash in your work? Explore Radar to follow supported changes and choose your alerts.
Radar
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-flashGoogle Gemini 3.7 Flash API documentation
- Context window
- 1MGoogle Gemini 3.7 Flash API documentation
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, image, video, audio, pdfGoogle Gemini 3.7 Flash API documentation
- Output modalities
- textGoogle Gemini 3.7 Flash API documentation
- Parameters
- Not disclosed by the provider
- Availability
- Gemini API · Google AI Studio · Gemini App - Spark · Gemini Enterprise App · Gemini Enterprise Agent Platform · Google AntigravityGoogle DeepMind Gemini 3.7 Flash model card
- Cloud regions
- Not tracked yet
- Lifecycle
- activeGoogle Gemini 3.7 Flash API documentation
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.075 per million cached input tokensGoogle Gemini API pricing
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
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 #19 of 194 on the public leaderboard with a score of 67.66/100. Its source-verified position is #14 of 74.
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.
Generally available as the stable `gemini-3.7-flash` model through the Gemini API and Google AI Studio. Google also distributes it through Gemini App - Spark, Gemini Enterprise App, Gemini Enterprise Agent Platform, and Google Antigravity.
Its explicit predecessor is Gemini 3.6 Flash. 25 of 486 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Multimodal & Grounded at #10, while its lowest eligible position is Agentic at #20. particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Last updated September 27, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does Gemini 3.7 Flash perform overall in AI benchmarks?
Gemini 3.7 Flash ranks #19 out of 194 models on the public BenchAlign leaderboard, with a score of 67.66/100. Its evidence status is Supported, and this profile shows 25 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Gemini 3.7 Flash good for knowledge and understanding?
Gemini 3.7 Flash ranks #10 out of 158 eligible models for knowledge and understanding, with a public category score of 71.1/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Gemini 3.7 Flash good for coding and programming?
Gemini 3.7 Flash ranks #17 out of 135 eligible models for coding and programming, with a public category score of 60.3/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Gemini 3.7 Flash good for reasoning and logic?
Gemini 3.7 Flash has source-displayable benchmark coverage for reasoning and logic, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is Gemini 3.7 Flash good for agentic tool use and computer tasks?
Gemini 3.7 Flash ranks #20 out of 105 eligible models for agentic tool use and computer tasks, with a public category score of 58.6/100. Higher-ranked alternatives are available for workloads where this category decides the choice. Check the underlying rows before treating the aggregate as a workload guarantee.
Is Gemini 3.7 Flash good for multimodal and grounded tasks?
Gemini 3.7 Flash ranks #10 out of 50 eligible models for multimodal and grounded tasks, with a public category score of 82.6/100. That places it in the current top ten for this category. Check the underlying rows before treating the aggregate as a workload guarantee.
Does Gemini 3.7 Flash have full benchmark coverage on BenchLM?
No. Gemini 3.7 Flash currently has 42 source-displayable rows across 486 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of Gemini 3.7 Flash?
Gemini 3.7 Flash has a documented context window of 1M. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.