Capability
70.1/100
field median 56.3
#16 of 230 ranked models
Gemini 3.1 Pro
Decision reading
Gemini 3.1 Pro scores 70.1 out of 100 and ranks #16 of 230. This profile shows 28 source-displayable benchmark rows; its strongest eligible category is Multimodal & Grounded at #12. API pricing is $2 input and $12 output per million tokens, with cached input at $0.2.
Released Feb 19, 2026 — see all recent releases
Data as of September 18, 2026 · How the score is built
Gemini 3.1 Pro will be repriced, updated or retired. Get each notice with its source and date. Follow model changes
Share or export
Strongest published evidence
Multimodal & Grounded ranks #12. Particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Validate before choosing
28 published rows leave some tracked benchmark slots empty. Agentic is its lowest eligible category at #121.
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.
Price
$2input / $12 output
input median $0.95
cached $0.20 · batch + cache $0.20 · blended $7
Speed
116tok/s
field median 90 tok/s
First token 26.96 s
Context
1Mtokens
field median 256,000
Maximum output length is tracked separately
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.
- Agentic4/6 verified
- Coding4/5 verified
- Reasoning2/2 verified
- Knowledge2/6 verified
- Math2/2 verified
- MultilingualNot measured
- Multimodal0/7 verified
- Inst. FollowingNot measured
Verified sourceProvisionalNot measured
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 #121 of 154Percentile 22ndWeight 22%6 benchmarksMixed sources | 39.7 | #121 of 154 | 22nd | 22% | 6 benchmarks | Mixed sources |
| CodingRank #88 of 154Percentile 43rdWeight 20%5 benchmarksMixed sources | 46.1 | #88 of 154 | 43rd | 20% | 5 benchmarks | Mixed sources |
| ReasoningRank Not rankedWeight 17%2 benchmarksVerified | 50.7 | Not ranked | Not available | 17% | 2 benchmarks | Verified |
| KnowledgeRank #22 of 184Percentile 89thWeight 12%6 benchmarksMixed sources | 66.0 | #22 of 184 | 89th | 12% | 6 benchmarks | Mixed sources |
| MathRank Not rankedWeight 5%2 benchmarksVerified | 54.2 | Not ranked | Not available | 5% | 2 benchmarks | Verified |
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | Not ranked | Not available | 7% | 0 benchmarks | Not measured |
| MultimodalRank #12 of 48Percentile 77thWeight 12%7 benchmarksReported | 79.1 | #12 of 48 | 77th | 12% | 7 benchmarks | Reported |
| Inst. FollowingWeight 5%0 benchmarksNot measured | 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.
Coding5 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| LiveCodeBench (Vals)LiveCodeBench, Vals AI run | Score88.5% | Versus best verified row Best verified: Claude Fable 5.1 · 90.5% | Gap2 behind | WeightWeighted 15% | |
| LiveCodeBench Pro | Score82.9% | Versus best verified row Best verified: Sakana Fugu-Ultra · 90.8% | Gap7.9 behind | WeightDisplay only | Secondary exact |
| React Native Evals | Score78.9% | Versus best verified row Best verified: Composer 2 · 96.1% | Gap17.2 behind | WeightDisplay only | Benchmark exact |
| Vibe Code BenchVibe Code Bench v1.1 | Score32.03% | Versus best verified row Best verified: Claude Opus 4.7 · 71.00% | Gap39 behind | WeightDisplay only | Benchmark exact |
| SWE-bench (Vals)SWE-bench, Vals AI run | Score78.8% | Versus best verified row Best verified: Claude Opus 5 · 97.0% | Gap18.2 behind | WeightDisplay only | Verified |
Agentic6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| Claw-Eval | Score57.8% | Versus best verified row Best verified: Ornith-1.5-397B · 81.4% | Gap23.6 behind | WeightDisplay only | Benchmark exact |
| DeepSearchQA | Score69.7% | Versus best verified row Best verified: Atria Dawn Preview · 96.0% | Gap26.3 behind | WeightDisplay only | Secondary exact |
| τ²-bench resultsτ²-Bench Tool-Agent-User Evaluation | Score95.6% | Versus best verified row Best verified: GPT-5.4 · 98.9% | Gap3.3 behind | WeightDisplay only | Secondary exact |
| Gert LabsGert Labs Composite Game Benchmark | Score56.87% | Versus best verified row Best verified: Claude Opus 4.8 · 72.97% | Gap16.1 behind | WeightDisplay only | Benchmark exact |
| ResearchClawBench | Score13.3% | Versus best verified row Best verified: Claude Opus 4.8 · 21.1% | Gap7.8 behind | WeightDisplay only | Benchmark exact |
| Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run | Score70.8% | Versus best verified row Best verified: GPT-6 Astra · 87.3% | Gap16.5 behind | WeightDisplay only |
Reasoning2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| ARC-AGI-2Abstraction and Reasoning Corpus for AGI v2 | Score77.1% | Versus best verified row Best verified: GPT-6 Astra · 95% | Gap17.9 behind | WeightWeighted 25% | Provider exact |
| ARC-AGI-3Abstraction and Reasoning Corpus for AGI v3 | Score0.4% | Versus best verified row Best verified: GPT-6 Astra · 62.7% | Gap62.3 behind | WeightWeighted 15% | Benchmark exact |
Knowledge6 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| HLE w/o toolsHumanity's Last Exam without tools | Score45.4% | Versus best verified row Best verified: Claude Fable 5.1 · 60.9% | Gap15.5 behind | WeightWeighted 10% | Secondary exact |
| MMLU-Pro (Vals)MMLU-Pro, Vals AI run | Score91.0% | Versus best verified row Best verified: Claude Fable 5.1 · 92.4% | Gap1.4 behind | WeightWeighted 10% | Verified |
| GPQA-DGPQA Diamond | Score94.3% | Versus best verified row Best verified: GPT-6 Astra · 96.0% | Gap1.7 behind | WeightDisplay only | Secondary exact |
| HealthBench Hard | Score20.6% | Versus best verified row Best verified: Muse Spark · 42.8% | Gap22.2 behind | WeightDisplay only | Secondary exact |
| MedXpertQA (Text)MedXpertQA Text | Score71.5% | Versus best verified row Best verified: Muse Spark · 52.6% | Gap18.9 behind | WeightDisplay only | Secondary exact |
| GPQA Diamond (Vals)GPQA Diamond, Vals AI run | Score95.5% | Versus best verified row Best verified: Gemini 3.1 Pro · 95.5% | GapBest verified | WeightDisplay only |
Math2 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)FrontierMath v2 Tiers 1-3 | Score36.900% | Versus best verified row Best verified: GPT-5.6 Sol · 89.000% | Gap52.1 behind | WeightWeighted 30% | Benchmark exact |
| FrontierMath v2 (Tier 4)FrontierMath v2 Tier 4 | Score16.700% | Versus best verified row Best verified: GPT-6 Astra · 97.600% | Gap80.9 behind | WeightWeighted 10% | Benchmark exact |
Multimodal7 rows
| Benchmark | Score | Versus best verified row | Gap | Weight | Evidence |
|---|---|---|---|---|---|
| MMMU-ProMassive Multi-discipline Multimodal Understanding Pro | Score83.9% | Versus best verified row Best verified: Gemini 3.5 Flash · 83.6% | Gap0.3 behind | WeightWeighted 40% | Secondary exact |
| CharXivCharXiv Reasoning | Score80.2% | Versus best verified row Best verified: Qwen3.8 Max · 93.5% | Gap13.3 behind | WeightWeighted 20% | Secondary exact |
| ERQA | Score69.4% | Versus best verified row Best verified: Qwen3.8 Max · 77.8% | Gap8.4 behind | WeightDisplay only | Secondary exact |
| SimpleVQA | Score72.4% | Versus best verified row Best verified: Qwen3.7 Plus · 81.7% | Gap9.3 behind | WeightDisplay only | Secondary exact |
| ScreenSpot Pro | Score84.4% | Versus best verified row Best verified: GPT-6 Astra · 92.7% | Gap8.3 behind | WeightDisplay only | Secondary exact |
| ZeroBench | Score29.0% | Versus best verified row Best verified: Muse Spark · 33.0% | Gap4 behind | WeightDisplay only | Secondary exact |
| MedXpertQA (MM)MedXpertQA Multimodal | Score81.3% | Versus best verified row Best verified: Qwen3.8 Max · 80.4% | Gap0.9 behind | WeightDisplay only | Secondary exact |
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 3.1 Pro category percentile values
- Agentic22nd percentile
- Coding43rd percentile
- ReasoningNot eligible
- Knowledge89th percentile
- MathNot eligible
- MultilingualNot eligible
- Multimodal77th percentile
- Instruction followingNot eligible
The dashed outline is median of 6 nearest peers.
Eligible category ranks
- Agentic#121/154
- Coding#88/154
- ReasoningNot ranked
- Knowledge#22/184
- MathNot ranked
- MultilingualNot ranked
- Multimodal#12/48
- Inst. FollowingNot ranked
Top decileTop quartileMid-fieldNot 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.
The chart opens on the current model. Scroll horizontally to inspect the full price axis.
Current modelGemini 3.1 Pro · 70.1 score · $7 blended per million tokens
Horizontal: blended price per million tokens, log scale · Vertical: public score
Published price history
This chart appears only when at least two dated first-party price records exist for the exact model. It shows rate changes, not an inferred cost trend.
2 dated first-party price records. The newest record lists $12 per million tokens.
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-3.1-pro-previewGoogle Gemini 3.1 Pro Preview model documentation
- Context window
- 1MGoogle Gemini 3.1 Pro Preview model documentation
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- text, image, video, audio, pdfGoogle Gemini 3.1 Pro Preview model documentation
- Output modalities
- textGoogle Gemini 3.1 Pro Preview model documentation
- Parameters
- Not disclosed by the provider
- Availability
- Gemini API · Google AI StudioGoogle Gemini model catalog
- Cloud regions
- Not tracked yet
- Lifecycle
- previewGoogle Gemini 3.1 Pro Preview model documentation
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.20 per million cached input tokensGoogle Gemini API pricing
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
Lineage
The sequence follows explicit supersedes links. A successor's displayed score stays at least 0.1 points above its predecessor; raw benchmark rows do not move. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Feb 19, 2026 · you are here
Gemini 3.1 ProScore 70.1 · $2 / $12
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 3.1 Pro ranks #16 of 230 on the public leaderboard with a score of 70.14/100. Its source-verified position is #13 of 130.
Gemini 3.1 Pro 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.
28 of 446 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Its strongest eligible category is Multimodal & Grounded at #12, while its lowest eligible position is Agentic at #121. particularly strong for screenshots, documents, charts, and grounded multimodal workflows.
Radar
Full release historyGemini 3.1 Pro release history
Gemini 3.1 Pro released
Google · Model release
Source confirmed
Radar confirmed these at the source. Use Gemini 3.1 Pro in your work? Explore Radar to follow supported changes and choose your alerts.
Questions
How does Gemini 3.1 Pro perform overall in AI benchmarks?
Gemini 3.1 Pro ranks #16 out of 230 models on the public BenchAlign leaderboard, with a score of 70.14/100. Its evidence status is Supported, and this profile shows 28 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.
Is Gemini 3.1 Pro good for knowledge and understanding?
Gemini 3.1 Pro ranks #22 out of 184 eligible models for knowledge and understanding, with a public category score of 66/100. 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 3.1 Pro good for coding and programming?
Gemini 3.1 Pro ranks #88 out of 154 eligible models for coding and programming, with a public category score of 46.1/100. 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 3.1 Pro good for mathematics?
Gemini 3.1 Pro 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 3.1 Pro good for reasoning and logic?
Gemini 3.1 Pro 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 3.1 Pro good for agentic tool use and computer tasks?
Gemini 3.1 Pro ranks #121 out of 154 eligible models for agentic tool use and computer tasks, with a public category score of 39.7/100. 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 3.1 Pro good for multimodal and grounded tasks?
Gemini 3.1 Pro ranks #12 out of 48 eligible models for multimodal and grounded tasks, with a public category score of 79.1/100. 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.
Does Gemini 3.1 Pro have full benchmark coverage on BenchLM?
No. Gemini 3.1 Pro currently has 46 source-displayable rows across 446 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 3.1 Pro?
Gemini 3.1 Pro 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.
Compare Gemini 3.1 Pro with every tracked model490 comparisons
Last updated September 18, 2026. Runtime fields remain blank until a sourced snapshot exists.
Watch Gemini 3.1 Pro in the weekly brief
Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.
Read a sample issueJoin 2,000+ readers.