Data as of November 17, 2026 · How the score is built
Gemini 3.5 Flash
Decision readingGemini 3.5 Flash scores 61.9 out of 100 and ranks #44 of 216. This profile shows 39 source-displayable benchmark rows; its strongest eligible category is Multimodal & Grounded at #15. API pricing is $1.5 input and $9 output per million tokens, with cached input at $0.15.
Email me when its price, rank or availability changes.
Capability
61.9/100
field median 49.4#44 of 216 ranked models
Public
#44of 216
Verified #28 of 80
Price
$1.50input / $9 output
input median $0.60cached $0.15 · batch + cache $0.075 · blended $5.25
Speed
188tok/s
field median 89.5 tok/sFirst token 12.96 s
Context
1Mtokens
field median 256,000Maximum output length is tracked separately
Decision snapshot
Where this model sits in the ranked field, its strongest published evidence, and what to validate before choosing it. The marker compares it with the current ranked catalog, not with an absolute quality threshold.
Strongest published evidence
Multimodal & Grounded ranks #15. Particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Validate before choosing
39 published rows leave some tracked benchmark slots empty. Agentic is its lowest eligible category at #57.
Source-linked · 39 displayable benchmark rows
Is Gemini 3.5 Flash right for your job?
Its place on the LLM Selector’s job rankings, the same order the job pages publish. Describe your own job to see where it stands for that.
Gemini 3.5 Flash has a newer version, Gemini 4 Hydrogen. Where Gemini 4 Hydrogen ranks by benchmark score:
We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy
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 | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticRank #57 of 138Percentile 59thWeight 22%9 benchmarksVerified | 50.2 | 9 benchmarks | Verified | |||
| CodingRank #46 of 163Percentile 72ndWeight 20%7 benchmarksVerified | 50.2 | 7 benchmarks | Verified | |||
| ReasoningRank #53 of 160Percentile 67thWeight 17%4 benchmarksVerified | 52.4 | 4 benchmarks | Verified | |||
| MultimodalRank #15 of 67Percentile 79thWeight 12%6 benchmarksVerified | 67.9 | 6 benchmarks | Verified | |||
| KnowledgeRank #27 of 181Percentile 86thWeight 12%4 benchmarksMixed sources | 65.6 | 4 benchmarks | Mixed sources | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingRank #39 of 157Percentile 76thWeight 5%Not listedUnlisted sources | 87.7 | Not listed | Unlisted sources | |||
| MathRank Not rankedWeight 5%2 benchmarksVerified | 62.6 | 2 benchmarks | Verified |
39 of 922 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.
- Agentic9/9 verified
- Coding7/7 verified
- Reasoning4/4 verified
- Multimodal6/6 verified
- Knowledge3/4 verified
- MultilingualNot measured
- Inst. FollowingNot measured
- Math2/2 verified
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 3.5 Flash category percentile values
- Agentic59th percentile
- Coding72nd percentile
- Reasoning67th percentile
- Multimodal79th percentile
- Knowledge86th percentile
- MultilingualNot eligible
- Instruction following76th percentile
- MathNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#57/138
- Coding#46/163
- Reasoning#53/160
- Multimodal#15/67
- Knowledge#27/181
- MultilingualNot ranked
- Inst. Following#39/157
- MathNot ranked
Benchmark ledger
Every result says who ran the test. For this model, Google ran 13 of 39. Coding opens by default. The tick marks the best source-verified result in the catalog; provisional leaders do not set the reference.
Coding7 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| SWE-bench Pro | 55.1% | Best verified: Claude Fable 5.2 · 93% | Gap37.9 behind | Reported by Google | Counts26% ref. weight |
| LiveCodeBench (Vals)LiveCodeBench, Vals AI run | 87.6% | Best verified: Claude Fable 5.1 · 90.5% | Gap2.9 behind | Independent run: Vals AI | Counts8% ref. weight |
| Terminal-Bench 2.1 | 76.2% | Best verified: SWE-2 · 92.8% | Gap16.6 behind | Reported by Google | CountsScored in Agentic |
| Vibe Code BenchVibe Code Bench v1.1 | 48.68% | Best verified: Gemini 4 Argon · 91.90% | Gap43.2 behind | Run by Vibe Code Bench | Shown only |
| cursorBench31 | 49.8% | Best verified: Claude Fable 5 · 70.6% | Gap20.8 behind | Run by cursorBench31 | Shown only |
| CursorBench 3.2 | 48.8% | Best verified: Claude Fable 5.1 · 73.4% | Gap24.6 behind | Run by CursorBench 3.2 | Shown only |
| SWE-bench (Vals)SWE-bench, Vals AI run | 78.8% | Best verified: Claude Opus 5 · 97.0% | Gap18.2 behind | Independent run: Vals AI | Shown only |
Agentic9 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| Terminal-Bench 2.1 | 76.2% | Best verified: SWE-2 · 92.8% | Gap16.6 behind | Reported by Google | Counts9% ref. weight |
| OSWorld-Verified | 78.4% | Best verified: Qwen3.8 Max · 86.1% | Gap7.7 behind | Reported by Google | Counts8% ref. weight |
| MCP Atlas | 83.6% | Best verified: Muse Spark Open · 89.2% | Gap5.6 behind | Reported by Google | Counts5% ref. weight |
| Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run | 74.2% | Best verified: Claude Opus 5.5 · 87.6% | Gap13.4 behind | Independent run: Vals AI | Counts4% ref. weight |
| Toolathlon | 56.5% | Best verified: Muse Spark 1.1 · 75.6% | Gap19.1 behind | Reported by Google | Counts3% ref. weight |
| Finance Agent v2 | 57.9% | Best verified: Gemini 4 Argon · 65.4% | Gap7.5 behind | Run by Finance Agent v2 | Shown only |
| Gert LabsGert Labs Composite Game Benchmark | 61.85% | Best verified: Claude Opus 4.8 · 72.97% | Gap11.1 behind | Run by Gert Labs | Shown only |
| ResearchClawBench | 18.0% | Best verified: Claude Opus 4.8 · 21.1% | Gap3.1 behind | Run by ResearchClawBench | Shown only |
| MazeBenchMazeBench: visual spatial reasoning in a 3D open world | 0% | Best verified: GPT-6 Astra · 24% | Gap24 behind | Run by MazeBench | Shown only |
Reasoning4 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| MRCRv2 | 77.3% | Best verified: Sakana Fugu-Ultra · 93.6% | Gap16.3 behind | Reported by Google | Counts20% ref. weight |
| ARC-AGI-2Abstraction and Reasoning Corpus for AGI v2 | 72.1% | Best verified: GPT-6 Astra · 95% | Gap22.9 behind | Reported by Google | Counts15% ref. weight |
| MRCR 1M | 26.6% | Best verified: DeepSeek V4 Pro 0813 · 83.5% | Gap56.9 behind | Reported by Google | Shown only |
| ARC-AGI-1ARC-AGI-1 Semi-Private Evaluation | 92.50% | Best verified: Claude Fable 5 · 98.50% | Gap6 behind | Run by ARC-AGI-1 | Shown only |
Multimodal6 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| MMMU-ProMassive Multi-discipline Multimodal Understanding Pro | 83.6% | Best verified: GPT-6.1 Terra · 85.5% | Gap1.9 behind | Reported by Google | Counts28% ref. weight |
| CharXivCharXiv Reasoning | 84.2% | Best verified: Qwen3.8 Max · 93.5% | Gap9.3 behind | Reported by Google | Counts16% ref. weight |
| ParseBench-1ParseBench-1: document parsing for AI agents (LlamaIndex leaderboard, first release) | 73.22% | Best verified: Claude Opus 5.5 · 78.80% | Gap5.6 behind | Run by ParseBench-1 | Counts6% ref. weight |
| ParseBench-2ParseBench-2: severity-aware document parsing (LlamaIndex leaderboard) | 74.73% | Best verified: Claude Opus 5.5 · 85.75% | Gap11 behind | Run by ParseBench-2 | Shown only |
| ExtractBench (LlamaIndex)ExtractBench: schema-guided enterprise document extraction (LlamaIndex leaderboard) | 79.84% | Best verified: GPT-6.1 Sol · 95.31% | Gap15.5 behind | Run by ExtractBench (LlamaIndex) | Shown only |
| Blueprint-Bench 2 | 33.6% | Best verified: Gemini 4 Carbon · 58.6% | Gap25 behind | Reported by Google | Shown only |
Knowledge4 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| HLEHumanity's Last Exam | 40.2% | Best verified: Claude Fable 5.2 · 74.2% | Gap34 behind | Reported by Google | Counts44% ref. weight |
| MMLU-Pro (Vals)MMLU-Pro, Vals AI run | 89.5% | Best verified: Claude Fable 5.1 · 92.4% | Gap2.9 behind | Independent run: Vals AI | Counts6% ref. weight |
| GPQA Diamond (Vals)GPQA Diamond, Vals AI run | 92.7% | Best verified: Gemini 3.1 Pro · 95.5% | Gap2.8 behind | Independent run: Vals AI | Counts2% ref. weight |
| GPQA-DGPQA Diamond | 92.7% | Best verified: GPT-6 Astra · 96.0% | Gap3.3 behind | Reported by a third party | Shown only |
Math2 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)FrontierMath v2 Tiers 1-3 | 62.807% | Best verified: Claude Fable 5.2 · 95.088% | Gap32.3 behind | Run by FrontierMath v2 (Tiers 1-3) | CountsWeighted 30% |
| FrontierMath v2 (Tier 4)FrontierMath v2 Tier 4 | 26.829% | Best verified: Claude Fable 5.2 · 100.000% | Gap73.2 behind | Run by FrontierMath v2 (Tier 4) | CountsWeighted 10% |
External signals7 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| Game Arena Go | 756 | Best verified: Claude Fable 5.2 · 1877 | Gap1121 behind | Run by Game Arena Go | Shown only |
| SkillsBench official leaderboard (BenchFlow)SkillsBench v1.1 official leaderboard (BenchFlow) | 48.2% | Best verified: GPT-5.5 · 67.3% | Gap19.1 behind | Run by the benchmark | Shown only |
| NYT ConnectionsNYT Connections LLM Benchmark (extended version) | 92.6% | Best verified: Gemini 4 Carbon · 99.3% | Gap6.7 behind | Run by NYT Connections | Shown only |
| BabaIsBenchBabaIsBench (Baba Is You intro levels, pass@1) | 88% | Best verified: Claude Opus 5 · 100% | Gap12 behind | Run by BabaIsBench | Shown only |
| Chess PuzzlesChess Puzzles (Epoch AI) | 45.0% | Best verified: Claude Fable 5.2 · 79.0% | Gap34 behind | Run by Chess Puzzles | Shown only |
| Mystery Game PuzzlesMystery Game Puzzles (Epoch AI) | 32.0% | Best verified: Claude Fable 5.2 · 90.0% | Gap58 behind | Run by Mystery Game Puzzles | Shown only |
| EBR-benchEarthborne Rangers (EBR-bench) | 4.8% | Best verified: Claude Fable 5.2 · 90.5% | Gap85.7 behind | Run by EBR-bench | Shown only |
Outlined bars are lab-reported. Filled bars are benchmark-run or independent.
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.
Gemini 3.5 Flash · 61.9 score · $5.25 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
Gemini 3.5 Flash CyberSpec 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-3.5-flashGoogle Gemini API pricing
- Context window
- 1MGoogle Gemini API pricing
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, image, video, audio, pdfGemini API model page for gemini-3.5-flash
- Output modalities
- textGemini API model page for gemini-3.5-flash
- Parameters
- Not disclosed by the provider
- Availability
- Not sourced yet
- Cloud regions
- Not tracked yet
- Lifecycle
- Superseded
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.15 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 3.5 Flash ranks #44 of 216 on the public leaderboard with a score of 61.92/100. Its source-verified position is #28 of 80.
Gemini 3.5 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.
Gemini 3.5 Flash sits in the Gemini 3.5 Flash family with Gemini 3.5 Flash Cyber. Its explicit predecessor is Gemini 3 Flash. 39 of 922 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Multimodal & Grounded at #15, while its lowest eligible position is Agentic at #57. particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Last updated November 17, 2026. Runtime fields remain blank until a sourced snapshot exists.
Deployment options
Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.
Where to run Gemini 3.5 Flash by API
2 hosts publish a per-token price for it. DeepInfra has the lowest output rate, $9.00 per million tokens. Prices read from each host’s own pages on November 16, 2026.
| Host | Input /1M | Output /1M | Cached /1M | Context |
|---|---|---|---|---|
| DeepInfra | $1.50 | $9.00 | — | 1M |
| OpenRouter | $1.50 | $9.00 | $0.15 | 1.05M |
Questions
How does Gemini 3.5 Flash perform overall in AI benchmarks?
Gemini 3.5 Flash ranks #44 out of 216 models on the public BenchAlign leaderboard, with a score of 61.92/100. Its evidence status is Supported, and this profile shows 39 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Gemini 3.5 Flash good for knowledge and understanding?
Gemini 3.5 Flash ranks #27 out of 181 eligible models for knowledge and understanding, with a public category score of 65.6/100. 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 3.5 Flash good for coding and programming?
Gemini 3.5 Flash ranks #46 out of 163 eligible models for coding and programming, with a public category score of 50.2/100. 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 3.5 Flash good for mathematics?
Gemini 3.5 Flash has source-displayable benchmark coverage for mathematics, 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 3.5 Flash good for reasoning and logic?
Gemini 3.5 Flash ranks #53 out of 160 eligible models for reasoning and logic, with a public category score of 52.4/100. 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 3.5 Flash good for agentic tool use and computer tasks?
Gemini 3.5 Flash ranks #57 out of 138 eligible models for agentic tool use and computer tasks, with a public category score of 50.2/100. 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 3.5 Flash good for multimodal and grounded tasks?
Gemini 3.5 Flash ranks #15 out of 67 eligible models for multimodal and grounded tasks, with a public category score of 67.9/100. 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.
Which sibling models are related to Gemini 3.5 Flash?
Gemini 3.5 Flash belongs to the Gemini 3.5 Flash family. Related tracked variants include Gemini 3.5 Flash Cyber. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Gemini 3.5 Flash have full benchmark coverage on BenchLM?
No. Gemini 3.5 Flash currently has 63 source-displayable rows. 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 3.5 Flash?
Gemini 3.5 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.