GoogleGemini 3.1 Pro
Reasoning effort
Gemini 3.1 Pro is a Google proprietary reasoning model, released 19 Feb 2026. Its best configuration ranks #105 of 775 on the BenchLeader Index at 60.7 ±4.0, in the upper half. The ± is the point: 200 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (68) and lowest in composite (49). At $4.50 per million tokens blended it is among the most expensive ranked models. Output speed of 131 tokens per second puts it faster than most, with a first answer in 21.4 s. It has been measured at 5 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 19 Oct 2026.
- Blended price
- $4.50/M
- $2.00 in · $12.00 out
- Output speed
- 131 tok/s
- measured by Artificial Analysis
- First answer
- 21 s
- first token 2.90 s
- Context
- 1.0M
- Full answer
- 25 s
- median, reasoning included
- Cost per run
- $1.30
- one full Intelligence Index run
- Released
- 19 Feb 2026
- Overall index61
- Reasoning68
- Coding57
- Agents & tools57
- Maths61
- Knowledge64
- Instruction following68
- Human preference59
- Multimodal65
- Long context66
- Composite49
Versions
Google has shipped 10 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Gemini 3.1 Prothis page | 19 Feb 2026 | 60.7 | #105 |
| Gemini 3 Pro | 18 Nov 2025 | 60.5 | #110 |
| Gemini 2.5 Pro | 25 Mar 2025 | 53.9 | #260 |
| Gemini 2.0 Pro | 5 Feb 2025 | 47.0 | #460 |
| Gemini 1.5 Pro 002 | 24 Sept 2024 | 44.2 | #534 |
| Gemini 1.5 Pro | 15 May 2024 | 42.2 | #594 |
| Gemini 1.5 Pro 001 | 14 May 2024 | 45.5 | #504 |
| Gemini 1.0 Pro 001 | 13 Dec 2023 | – | – |
| Gemini 1.0 Pro | 6 Dec 2023 | – | – |
| Gemini 1.0 Pro 002 | – | – | – |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| low | – | – | 131 tok/s | 21 s | $0.0044 | |
| medium | – | – | 131 tok/s | 21 s | $0.0044 | |
| high | 60.3 | #113 | 131 tok/s | 21 s | $0.0044 | Agents & tools 58 · Coding 58 · Knowledge 66 · Maths 66 · Reasoning 65 |
| thinking | – | – | 131 tok/s | 21 s | $0.0044 | Coding 61 |
| not statedbest | 60.7 | #105 | 131 tok/s | 21 s | $0.0044 | Agents & tools 57 · Coding 57 · Composite 49 · Human preference 59 · Instruction following 68 · Knowledge 64 · Long context 66 · Maths 61 · Multimodal 65 · Reasoning 68 |
How its index has moved
8 Sept 2026 to 19 Oct 2026- not stated
- high
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|
| GPQA Diamond | – | – | 94.4%#8 | 94.1%#11 | Epoch AI Benchmarking Hub | |
| Humanity's Last Exameffort not stated | – | – | – | 46.4%#3 | Scale AI / CAIS | |
| SimpleBencheffort not stated | – | – | – | 79.6%#3 | SimpleBench | |
| Arena Hard Promptseffort not stated | – | – | – | 1508#19 | Arena (LMArena) | |
| LiveBench 2.0 Reasoningnot in indexeffort not stated | – | – | – | 84.0%#31 | LiveBench | |
| LiveBench 1.0 Reasoningnot in index | – | – | 84.0%#41 | – | LiveBench | |
| GPQA Diamond (AA)not in indexeffort not stated | – | – | – | 94.1%#7 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in indexeffort not stated | – | – | – | 47.0%#41 | Artificial Analysis | |
| GPQA Diamond (Vals)not in index | – | – | 95.5%#1 | – | Vals AI | |
| EnigmaEval | – | – | 36.8%#3 | – | Scale AI SEAL | |
| ARC-AGI-1effort not stated | – | – | – | 98.0%#11 | ARC Prize | |
| ARC-AGI-2effort not stated | – | – | – | 77.1%#43 | ARC Prize | |
| ARC-AGI-3effort not stated | – | – | – | 0.4%#46 | ARC Prize | |
| CritPteffort not stated | – | – | – | 17.7%#73 | Artificial Analysis | |
| MysteryMechanismnot in index | – | – | 20.7%#22 | – | Vals AI | |
| Chess Puzzlesnot in index | – | – | 51.0%#18 | – | Epoch AI Benchmarking Hub | |
| Mystery Game Puzzlesnot in index | 29.0%#35 | 32.0%#29 | 34.0%#24 | – | Epoch AI Benchmarking Hub | |
| CL-benchnot in indexeffort not stated | – | – | – | 20.8%#6 | Epoch AI Benchmarking Hub | |
| CL-bench Lifenot in indexeffort not stated | – | – | – | 16.9%#6 | Epoch AI Benchmarking Hub | |
| LMCAnot in index | 51.2%#60 | 52.8%#52 | 53.8%#50 | – | Epoch AI Benchmarking Hub | |
| DTBenchnot in index | 94.9%#46 | 97.1%#17 | 95.5%#38 | – | Epoch AI Benchmarking Hub | |
| ForecastBenchnot in index | – | – | 65.4%#13 | 64.3%#28 | ForecastBench |
Coding
| Benchmark | high | thinking | not stated | Source | Trend |
|---|---|---|---|---|---|
| SciCodeeffort not stated | – | – | 58.7%#20 | SciCode | |
| WeirdMLeffort not stated | – | – | 72.1%#31 | WeirdML | |
| GSO-Bencheffort not stated | – | – | 22.6%#16 | GSO-Bench | |
| Arena Codingeffort not stated | – | – | 1521#33 | Arena (LMArena) | |
| Arena WebDeveffort not stated | – | – | 1447#71 | Arena (LMArena) | |
| LiveBench 2.0 Codingnot in indexeffort not stated | – | – | 66.7%#38 | LiveBench | |
| LiveBench 1.0 Codingnot in index | 76.5%#45 | – | – | LiveBench | |
| SciCode (AA)not in indexeffort not stated | – | – | 58.7%#20 | Artificial Analysis | |
| LiveCodeBench | 88.5%#6 | – | – | Vals AI | |
| IOI | 51.8%#23 | – | – | Vals AI | |
| SWE-bench (Vals)not in index | 78.8%#28 | – | – | Vals AI | |
| SWE-Bench Pro | – | 46.1%#5 | – | Scale AI SEAL | |
| Code Migrationnot in index | 17.3%#53 | – | – | Vals AI | |
| ProgramBenchnot in index | 0.0%#32 | – | – | Vals AI | |
| Vibe Code Bench 1-100not in index | 6.7%#24 | – | – | Vals AI | |
| Vibe Code Bench v1.1not in index | 32.0%#70 | – | – | Vals AI | |
| SWE-Bench Pro (private)not in index | – | 46.1%#5 | – | Scale AI SEAL | |
| SWE Atlas: Codebase QnAnot in indexeffort not stated | – | – | 13.5%#19 | Scale AI SEAL | |
| SWE Atlas: Refactoringnot in indexeffort not stated | – | – | 33.8%#12 | Scale AI SEAL | |
| SWE Atlas: Test Writingnot in indexeffort not stated | – | – | 29.8%#18 | Scale AI SEAL | |
| MirrorCodenot in index | 8.9%#9 | – | – | Epoch AI Benchmarking Hub | |
| DeepSWE v1.1not in indexeffort not stated | – | – | 11.7%#67 | Epoch AI Benchmarking Hub | |
| ALE-Benchnot in indexeffort not stated | – | – | 1160.6#47 | Epoch AI Benchmarking Hub | |
| AlgoTunenot in indexeffort not stated | – | – | 2.0#2 | AlgoTune | |
| Terminal-Bench 4.0 (AA)not in indexeffort not stated | – | – | 4.0%#104 | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in indexeffort not stated | – | – | 73.8%#68 | Artificial Analysis |
Agents & tools
| Benchmark | low | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| Terminal-Bencheffort not stated | – | – | 80.2%#3 | Terminal-Bench | |
| APEX-Agentseffort not stated | – | – | 35.1%#46 | Mercor | |
| Arena Agenteffort not stated | – | – | -3.9#37 | Arena (LMArena) | |
| LiveBench Long Agenticnot in indexeffort not stated | – | – | 23.9%#37 | LiveBench | |
| LiveBench Browsernot in indexeffort not stated | – | – | 4.0%#42 | LiveBench | |
| LiveBench Agentic Codingnot in index | – | 44.1%#55 | – | LiveBench | |
| Terminal-Bench Hardeffort not stated | – | – | 53.8%#16 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in indexeffort not stated | – | – | 95.6%#19 | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals) | – | 70.8%#27 | – | Vals AI | |
| MCP Atlas | – | 78.2%#14 | – | Scale AI SEAL | |
| HiL-Bencheffort not stated | – | – | 35.3%#21 | Scale AI SEAL | |
| τ³-bench Bankingnot in index | – | 26.0%#16 | – | τ²-bench | |
| GDPval-AA v2.1not in indexeffort not stated | – | – | 14.6%#188 | Artificial Analysis | |
| τ³-Banking (AA)effort not stated | – | – | 21.4%#90 | Artificial Analysis | |
| ITBench SRE (AA)not in indexeffort not stated | – | – | 30.3%#33 | Artificial Analysis | |
| Analyst Agent (AA)not in indexeffort not stated | – | – | 41.3%#22 | Artificial Analysis | |
| APEX-Agents (AA)not in indexeffort not stated | – | – | 32.0%#11 | Artificial Analysis | |
| Finance Agent v2not in index | – | 43.0%#56 | – | Vals AI | |
| Legal Research Benchnot in index | – | 20.7%#56 | – | Vals AI | |
| SREBenchnot in index | – | 0.0%#31 | – | Vals AI | |
| Tax Agent Benchnot in index | – | 9.3%#56 | – | Vals AI | |
| Terminal-Bench 2.0 (Vals)not in index | – | 67.4%#4 | – | Vals AI | |
| Terminal-Bench 4.0 (Vals)not in index | – | 2.5%#44 | – | Vals AI | |
| Terminal-Bench Sciencenot in index | – | 1.4%#35 | – | Vals AI | |
| BALROGnot in indexeffort not stated | – | – | 57.0%#5 | BALROG | |
| DeepResearch Benchnot in index | 44.5%#31 | 47.8%#22 | – | DeepResearch Bench | |
| PostTrainBenchnot in indexeffort not stated | – | – | 22.0%#11 | Epoch AI Benchmarking Hub | |
| ExploitBenchnot in indexeffort not stated | – | – | 26.1%#4 | Epoch AI Benchmarking Hub | |
| METR Time Horizonsnot in indexeffort not stated | – | – | 77.0%#3 | METR | |
| GBAEvalnot in indexeffort not stated | – | – | 0.8%#18 | Epoch AI Benchmarking Hub | |
| Vending-Bench 2not in indexeffort not stated | – | – | 911.2#56 | Andon Labs | |
| AutomationBenchnot in indexeffort not stated | – | – | 35.4%#106 | Artificial Analysis | |
| GDP.pdf (AA)not in indexeffort not stated | – | – | 17.8%#77 | Artificial Analysis | |
| Harvey LAB-AA v1.1not in indexeffort not stated | – | – | 0.0%#23 | Harvey | |
| EnterpriseOps-Gymnot in indexeffort not stated | – | – | 42.2%#24 | Artificial Analysis | |
| AA-Briefcase v1.1not in indexeffort not stated | – | – | 457#55 | Artificial Analysis | |
| BrowseCompnot in indexeffort not stated | – | – | 31.2%#1 | OpenAI | |
| DeepSearchQAnot in indexeffort not stated | – | – | 60.2%#4 | ||
| FACTS Searchnot in indexeffort not stated | – | – | 83.7%#8 |
Maths
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3effort not stated | – | 59.6%#39 | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4effort not stated | – | 26.8%#42 | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 95.6%#43 | 95.6%#42 | Epoch AI Benchmarking Hub | |
| ProofBencheffort not stated | – | 26.0%#44 | Vals AI | |
| LiveBench Mathematicsnot in index | 91.0%#32 | – | LiveBench | |
| AIME (Vals) | 98.1%#1 | – | Vals AI | |
| AIME 2026effort not statedpossible contamination | – | 98.3%#4 | MathArena | |
| HMMT February 2026effort not statedpossible contamination | – | 94.7%#8 | MathArena | |
| USAMO 2026not in indexeffort not stated | – | 74.4%#3 | MathArena | |
| MathArena Apexeffort not statedpossible contamination | – | 60.9%#5 | MathArena | |
| Arena Mathseffort not stated | – | 1488#29 | Arena (LMArena) |
Knowledge
Instruction following
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| LiveBench Languagenot in index | 85.4%#14 | – | LiveBench | |
| LiveBench Instruction Followingnot in index | 79.1%#3 | – | LiveBench | |
| IFBencheffort not stated | – | 77.1%#18 | Artificial Analysis | |
| MultiChallengeeffort not stated | – | 71.4%#4 | Scale AI SEAL | |
| TutorBencheffort not stated | – | 53.0%#12 | Scale AI SEAL | |
| Arena Instruction Followingnot in indexeffort not stated | – | 1480#21 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1487#19 | Arena (LMArena) | |
| EQ-Bench 4 | 1142#20 | EQ-Bench | |
| Arena Creative Writingnot in index | 1481#12 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1496#13 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1500#15 | Arena (LMArena) |
Multimodal
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Arena Visioneffort not stated | – | 1296#22 | Arena (LMArena) | |
| LiveBench Designnot in indexeffort not stated | – | 61.6%#42 | LiveBench | |
| MMMU-Proeffort not stated | – | 82.4%#32 | Artificial Analysis | |
| MMMU-Pro (official)not in index | 80.5%#2 | – | MMMU | |
| MMMU-Pro (Vals)not in index | 88.2%#10 | – | Vals AI | |
| MortgageTaxnot in index | 69.4%#6 | – | Vals AI | |
| SAGEnot in index | 48.7%#26 | – | Vals AI | |
| VTBnot in indexeffort not stated | – | 29.0%#3 | Scale AI SEAL | |
| Blueprint-Bench 2not in indexeffort not stated | – | 26.5%#22 | Epoch AI Benchmarking Hub |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 82.0%#46 | Artificial Analysis | |
| Arena Documentnot in index | 1459#27 | Arena (LMArena) | |
| MLCRnot in index | 15.6%#45 | Artificial Analysis |
Composite
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | 155.3#35 | Epoch AI Benchmarking Hub | |
| LiveBench 2.0effort not stated | – | 54.9%#41 | LiveBench | |
| LiveBench 1.0not in index | 77.0%#26 | – | LiveBench | |
| AA Intelligence Index v4.3.2effort not stated | – | 29.7#125 | Artificial Analysis | |
| Vals Multimodal Indexnot in index | 56.1%#20 | – | Vals AI |
Safety & honesty
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| FORTRESSnot in index | 29.8%#25 | Scale AI SEAL | |
| MASKnot in indexpossible contamination | 42.4%#64 | Scale AI SEAL |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Google AI Studio | 113 tok/s | 2.73 s | $1.00 | $6.00 | 1.0M | – |
| Google AI Studio | 103 tok/s | 3.95 s | $2.00 | $12.00 | 1.0M | – |
| Google Vertex | 83 tok/s | 2.96 s | $2.00 | $12.00 | 1.0M | – |
| Google Vertex | 12 tok/s | 94 s | $1.00 | $6.00 | 1.0M | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.200 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0044 | $0.0039 | 23.7 s |
| Summarize a 30-page report | 12,000 / 600 | $0.031 | $0.015 | 26.0 s |
| Code edit | 6,000 / 1,500 | $0.030 | $0.022 | 32.9 s |
| Agentic coding session | 60,000 / 4,000 | $0.168 | $0.087 | 52.0 s |
| Structured extraction | 2,000 / 200 | $0.0064 | $0.0037 | 23.0 s |
More from Google
Compare Gemini 3.1 Pro with…
- Gemini 3.1 Pro vs Claude Opus 5.5
- Gemini 3.1 Pro vs Gemini 4 Argon
- Gemini 3.1 Pro vs Muse Wonder Full
- Gemini 3.1 Pro vs GPT-6 Astra
- Gemini 3.1 Pro vs Claude Fable 5.1
- Gemini 3.1 Pro vs GPT-6.1 Sol
- Gemini 3.1 Pro vs Claude Sonnet 5.5
- Gemini 3.1 Pro vs Claude Fable 5
- Gemini 3.1 Pro vs Claude Opus 5
- Gemini 3.1 Pro vs GPT-5.6 Sol
See also
Data as of 21 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Gemini 3.1 Pro: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/gemini-3-1-pro, data as of 21 Oct 2026.