Gemini 3.1 Pro vs Gemini 3.7 Flash
Full profiles: Gemini 3.1 Pro (Google) and Gemini 3.7 Flash (Google): every benchmark, price and rank.
Updated · Data from 14 sources
As of 10 Nov 2026, Gemini 3.1 Pro and Gemini 3.7 Flash are neck and neck on quality (60.5 vs 62.1 on the BenchLeader Index, inside the ±5.5 margin), and Gemini 3.7 Flash is 3.0× cheaper at $1.50 vs $4.50 per 1M tokens blended. Choose 3.1 Pro for agents & tools (57 vs 48), where it also streams faster; choose 3.7 Flash for fast first answers (2.5 s vs 13.7 s) and human preference (68 vs 58), knowledge (66 vs 56), and coding (63 vs 57).
Key facts (as of 10 Nov 2026)
- Gemini 3.1 Pro scores 60.5 on the BenchLeader Index and Gemini 3.7 Flash scores 62.1 (BenchLeader, 10 Nov 2026).
- Gemini 3.1 Pro costs $4.50 per 1M tokens blended (list price, 10 Nov 2026).
- Gemini 3.7 Flash costs $1.50 per 1M tokens blended (list price, 10 Nov 2026).
- Gemini 3.1 Pro streams faster than Gemini 3.7 Flash (Artificial Analysis, 7-day median, 10 Nov 2026).
- Of the 36 benchmarks both models have, Gemini 3.7 Flash leads 29 (BenchLeader, 10 Nov 2026).
- On Arena WebDev, Gemini 3.1 Pro is rated 1446 and Gemini 3.7 Flash 1588 (LMArena, as published by 10 Nov 2026).
Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Gemini 3.7 Flash | 63 vs 57 category score; Arena WebDev 1588 vs 1446 | Medium |
| Agents & tool use | Gemini 3.1 Pro | 57 vs 48 category score; Terminal-Bench 80.2% vs 12.4% | High |
| Maths | Gemini 3.7 Flash | 63 vs 61 category score; FrontierMath Tiers 1–3 71.6% vs 59.6% | Medium |
| Reasoning | Gemini 3.7 Flash | 68 vs 66 category score; GPQA Diamond 94.8% vs 94.1% | Medium |
| Knowledge | Gemini 3.7 Flash | 66 vs 56 category score; GDP.pdf 23.8% vs 17.0% | High |
| Long documents | Either | 66 vs 66 category score; AA-LCR v1.1 82.0% vs 81.7% | Either |
| Images and other media | Gemini 3.7 Flash | 67 vs 65 category score; Arena Vision 1313 vs 1296 | Medium |
| Chat quality (human preference) | Gemini 3.7 Flash | 68 vs 58 on Arena votes | High |
| Lowest cost | Gemini 3.7 Flash | $1.50 vs $4.50 per 1M blended; agentic coding session $0.060 vs $0.168 | – |
| Fastest first answer | Gemini 3.7 Flash | 2.5 s (Gemini 3.7 Flash (high)) vs 13.7 s (Gemini 3.1 Pro) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Gemini 3.1 Pro and Gemini 3.7 Flash (high) are neck and neck on quality (60.5 vs 62.1, within the ±5.5 margin).
- Gemini 3.1 Pro is stronger in agents & tools, instruction following, long context.
- Gemini 3.7 Flash (high) is stronger in coding, human preference, knowledge, maths, multimodal, reasoning.
- Gemini 3.7 Flash (high) is 3.0× cheaper ($1.50 vs $4.50 per 1M blended).
At a glance
| Spec | Gemini 3.1 Pro | Gemini 3.7 Flash |
|---|---|---|
| Developer | ||
| Released | 19 Feb 2026 | 13 Aug 2026 |
| Context window (tokens) | 1,048,576 | 1,048,576 |
| Compared at | Gemini 3.1 Pro | Gemini 3.7 Flash (high) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $2.00 / $12.00 / $0.200 | $0.750 / $3.75 / $0.075 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Gemini 3.1 Pro | Gemini 3.7 Flash (high) | Leader |
|---|---|---|---|
| Reasoning | 65.9 | 67.9 | Gemini 3.7 Flash |
| Coding | 56.7 | 62.7 | Gemini 3.7 Flash |
| Agents & tools | 57.1 | 48.0 | Gemini 3.1 Pro |
| Maths | 60.7 | 62.8 | Gemini 3.7 Flash |
| Knowledge | 56.2 | 66.1 | Gemini 3.7 Flash |
| Instruction following | 67.1 | – | not comparable |
| Human preference | 58.3 | 68.2 | Gemini 3.7 Flash |
| Multimodal | 64.5 | 67.2 | Gemini 3.7 Flash |
| Long context | 65.7 | 65.6 | level |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Instruction following.
Where sources disagree
Across the 36 benchmarks both have, Gemini 3.1 Pro leads 7, Gemini 3.7 Flash leads 29. Every overall index that rates both puts Gemini 3.7 Flash ahead.
| Overall index | Gemini 3.1 Pro | Gemini 3.7 Flash (high) | Leader |
|---|---|---|---|
| BenchLeader Index | 60.5 | 62.1 | neck and neck |
| Arena Text | 1487 | 1489 | Gemini 3.7 Flash |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Gemini 3.1 Pro | Gemini 3.7 Flash (high) |
|---|---|---|
| Blended price per 1M tokens | $4.50 | $1.50 |
| Input / output per 1M | $2.00 / $12.00 | $0.750 / $3.75 |
| Cached input per 1M | $0.200 | $0.075 |
| Price per task: chat reply | $0.0044 | $0.0014 |
| Price per task: summarize a 30-page report | $0.031 | $0.011 |
| Price per task: code edit | $0.030 | $0.010 |
| Price per task: agentic coding session | $0.168 | $0.060 |
| Price per task: structured extraction | $0.0064 | $0.0023 |
| Agentic coding session, input mostly cached | $0.087 | $0.030 |
| Output speed | 108 tok/s | 86 tok/s |
| Time to first answer | 13.7 s | 2.5 s |
| Time to a complete answer | 35.1 s | – |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.1 Pro scores 60.5 and takes 14 s to its first answer token; Gemini 3.1 Pro (high) scores 60.1 in 14 s. Gemini 3.7 Flash (high) scores 62.1 and takes 2 s to its first answer token; Gemini 3.7 Flash (medium) scores 60.2 in 2 s, Gemini 3.7 Flash (low) scores 57.4 in 2 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
155 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Gemini 3.1 Pro | Gemini 3.7 Flash (high) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 60.5 | 62.1 |
| Agents & tools score | BenchLeader | 57.1 | 48.0 |
| Coding score | BenchLeader | 56.7 | 62.7 |
| Human preference score | BenchLeader | 58.3 | 68.2 |
| Instruction following score | BenchLeader | 67.1 | – |
| Knowledge score | BenchLeader | 56.2 | 66.1 |
| Long context score | BenchLeader | 65.7 | 65.6 |
| Maths score | BenchLeader | 60.7 | 62.8 |
| Multimodal score | BenchLeader | 64.5 | 67.2 |
| Reasoning score | BenchLeader | 65.9 | 67.9 |
| Blended price $/M | BenchLeader | $4.50 | $1.50 |
| Output speed | BenchLeader | 108 tok/s | 86 tok/s |
| Time to first answer | BenchLeader | 13.7 s | 2.5 s |
| Context window | BenchLeader | 1,048,576 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 94.1% | 94.8% |
| Humanity's Last Exam | Scale AI / CAIS | 46.4% | – |
| SimpleBench | SimpleBench | 79.6% | – |
| Arena Hard Prompts | Arena (LMArena) | 1507 | 1510 |
| LiveBench 1.0 Reasoning | LiveBench | 84.0%† | 87.8% |
| GPQA Diamond (AA) | Artificial Analysis | 94.1% | 94.5% |
| Humanity's Last Exam (AA) | Artificial Analysis | 47.0% | 47.9% |
| HLE-Diamond (AA) | Artificial Analysis | 31.4% | – |
| GPQA Diamond (Vals) | Vals AI | 95.5%† | 93.9% |
| EnigmaEval | Scale AI SEAL | 36.8%† | – |
| ARC-AGI-1 | ARC Prize | 98.0% | 95.5% |
| ARC-AGI-2 | ARC Prize | 77.1% | 84.6% |
| ARC-AGI-3 | ARC Prize | 0.4% | – |
| CritPt | Artificial Analysis | 17.7% | 14.3% |
| MysteryMechanism | Vals AI | 20.7%† | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 51.0%† | 66.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 34.0%† | 37.0% |
| CL-bench | Epoch AI Benchmarking Hub | 20.8% | – |
| CL-bench Life | Epoch AI Benchmarking Hub | 16.9% | – |
| LMCA | Epoch AI Benchmarking Hub | 53.8%† | 50.4% |
| DTBench | Epoch AI Benchmarking Hub | 95.5%† | 96.8% |
| ForecastBench | ForecastBench | 64.1% | 66.2% |
| Coding | |||
| SciCode | SciCode | 58.7% | 57.2% |
| WeirdML | WeirdML | 72.1% | – |
| FrontierCode | Cognition | – | 43.6%† |
Show all 155 benchmarks
| Metric | Publisher | Gemini 3.1 Pro | Gemini 3.7 Flash (high) |
|---|---|---|---|
| Coding | |||
| GSO-Bench | GSO-Bench | 22.6% | – |
| Arena Coding | Arena (LMArena) | 1520 | 1521 |
| Arena WebDev | Arena (LMArena) | 1446 | 1588 |
| LiveBench 1.0 Coding | LiveBench | 76.5%† | 78.9% |
| LiveCodeBench | Vals AI | 88.5%† | 88.7% |
| IOI | Vals AI | 51.8%† | 67.8% |
| SWE-bench (Vals) | Vals AI | 78.8%† | 80.8% |
| SWE-Bench Pro | Scale AI SEAL | 46.1%† | – |
| Code Migration | Vals AI | 17.3%† | 34.8% |
| ProgramBench | Vals AI | 0.0%† | 0.0% |
| Vibe Code Bench 1-100 | Vals AI | 6.7%† | – |
| Vibe Code Bench v1.1 | Vals AI | 32.0%† | 70.4% |
| SWE-Bench Pro (private) | Scale AI SEAL | 46.1%† | – |
| SWE Atlas: Codebase QnA | Scale AI SEAL | 13.5% | – |
| SWE Atlas: Refactoring | Scale AI SEAL | 33.8% | – |
| SWE Atlas: Test Writing | Scale AI SEAL | 29.8% | – |
| MirrorCode | Epoch AI Benchmarking Hub | 8.9%† | 41.1% |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 11.7% | 65.3% |
| ALE-Bench | Epoch AI Benchmarking Hub | 1160.6 | 904.3 |
| AlgoTune | AlgoTune | 2.0 | – |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 20.3% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 4.0% | 13.6% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 73.8% | 85.8% |
| Design Arena Website | Design Arena | 1230 | 1300† |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 80.2% | 12.4% |
| Remote Labor Index | Scale AI / CAIS | – | 5.0%† |
| APEX-Agents | Mercor | 35.1% | 67.8%† |
| Arena Agent | Arena (LMArena) | -4.3 | -1.9 |
| LiveBench Agentic Coding | LiveBench | 44.1%† | 58.3% |
| Terminal-Bench Hard | Artificial Analysis | 53.8% | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 95.6% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 70.8%† | 77.5% |
| MCP Atlas | Scale AI SEAL | 78.2%† | – |
| HiL-Bench | Scale AI SEAL | 35.3% | – |
| τ³-bench Banking | τ²-bench | 26.0%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 14.6% | – |
| GDPval-AA v2.1 | Artificial Analysis | 799 | – |
| τ³-Banking (AA) | Artificial Analysis | 21.4% | 32.8% |
| ITBench SRE (AA) | Artificial Analysis | 30.3% | – |
| Analyst Agent (AA) | Artificial Analysis | 41.3% | 60.0% |
| APEX-Agents (AA) | Artificial Analysis | 32.0% | – |
| CyberBench | Vals AI | – | 43.8% |
| Finance Agent v2 | Vals AI | 43.0%† | 59.0% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 0.0%† | 8.8% |
| Legal Research Bench | Vals AI | 20.7%† | 34.6% |
| SkillsBench | Vals AI | – | 65.9% |
| SREBench | Vals AI | 0.0%† | 4.6% |
| Tax Agent Bench | Vals AI | 9.3%† | 22.8% |
| Terminal-Bench 2.0 (Vals) | Vals AI | 67.4%† | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | 2.5%† | 12.1% |
| Terminal-Bench Science | Vals AI | 1.4%† | 7.1% |
| Arena Search | Arena (LMArena) | 1209 | – |
| BALROG | BALROG | 57.0% | – |
| DeepResearch Bench | DeepResearch Bench | 47.8%† | – |
| PostTrainBench | Epoch AI Benchmarking Hub | 22.0% | – |
| ExploitBench | Epoch AI Benchmarking Hub | 26.1% | – |
| METR Time Horizons | METR | 77.0% | – |
| GBAEval | Epoch AI Benchmarking Hub | 0.8% | – |
| Vending-Bench 2 | Andon Labs | 911.2 | – |
| AutomationBench | Artificial Analysis | 35.4% | 62.0% |
| GDP.pdf (AA) | Artificial Analysis | 17.8% | 23.6% |
| Harvey LAB-AA v1.1 | Harvey | 0.0% | – |
| EnterpriseOps-Gym | Artificial Analysis | 42.2% | 50.4%† |
| AA-Briefcase v1.1 | Artificial Analysis | 457 | – |
| Design Arena Agentic | Design Arena | 1080 | 1216† |
| BrowseComp | OpenAI | 31.2% | – |
| DeepSearchQA | 60.2% | – | |
| FACTS Search | 90.2% | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 59.6% | 71.6% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 26.8% | 36.6% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 95.6% | 97.2% |
| ProofBench | Vals AI | 26.0% | 58.0%† |
| LiveBench Mathematics | LiveBench | 91.0%† | 93.5% |
| AIME (Vals) | Vals AI | 98.1%† | – |
| AIME 2026 | MathArena | 98.3% | – |
| HMMT February 2026 | MathArena | 94.7% | – |
| USAMO 2026 | MathArena | 74.4% | – |
| MathArena Apex | MathArena | 60.9% | – |
| Arena Maths | Arena (LMArena) | 1488 | 1502 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 73.5%† | 69.2% |
| LiveBench 1.0 Data Analysis | LiveBench | 78.5%† | 68.0% |
| MMLU-Pro | Vals AI | 91.0%† | 90.1% |
| LegalBench | Vals AI | 87.4%† | 87.3% |
| CorpFin | Vals AI | 64.5%† | – |
| TaxEval | Vals AI | 72.9%† | 74.7% |
| MedQA | Vals AI | 96.4%† | – |
| PRBench Finance | Scale AI SEAL | 41.9% | – |
| PRBench Legal | Scale AI SEAL | 44.0% | – |
| MultiNRC | Scale AI SEAL | 64.7% | – |
| BioMysteryBench | Vals AI | 60.0%† | – |
| CaseLaw v2 | Vals AI | 64.8%† | – |
| Excel Modeling Benchmark | Vals AI | 52.6%† | 71.3% |
| MedCode | Vals AI | 59.1%† | 53.4% |
| MedScribe | Vals AI | 76.1%† | 83.9% |
| Public Benefits Bench v1.1 | Vals AI | 53.8%† | – |
| DrugDiscoveryBench | Scale AI SEAL | 41.9%† | – |
| EBR-bench | Epoch AI Benchmarking Hub | 14.3% | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 17.0% | 23.8% |
| Global-MMLU-Lite (AA) | Artificial Analysis | 93.2% | – |
| Instruction following | |||
| LiveBench Language | LiveBench | 85.4%† | 85.5% |
| LiveBench Instruction Following | LiveBench | 79.1%† | 79.9% |
| IFBench | Artificial Analysis | 77.1% | – |
| MultiChallenge | Scale AI SEAL | 71.4% | – |
| TutorBench | Scale AI SEAL | 53.0% | – |
| Arena Instruction Following | Arena (LMArena) | 1480 | 1486 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1487 | 1489 |
| EQ-Bench 4 | EQ-Bench | 1132 | – |
| Arena Creative Writing | Arena (LMArena) | 1482 | 1487 |
| Arena Multi-turn | Arena (LMArena) | 1495 | 1490 |
| Arena Longer Queries | Arena (LMArena) | 1499 | 1501 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1296 | 1313 |
| MMMU-Pro | Artificial Analysis | 82.4% | 85.5% |
| MMMU-Pro (official) | MMMU | 80.5%† | – |
| MMMU-Pro (Vals) | Vals AI | 88.2%† | 89.0% |
| MortgageTax | Vals AI | 69.4%† | 66.7% |
| SAGE | Vals AI | 83.2%† | – |
| VTB | Scale AI SEAL | 29.0% | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 26.5% | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 82.0% | 81.7% |
| Arena Document | Arena (LMArena) | 1459 | 1468 |
| MLCR | Artificial Analysis | 11.1% | 12.2% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 156.3 | 158.4† |
| LiveBench 1.0 | LiveBench | 77.0%† | 78.8% |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 29.7 | 39.1 |
| AA Progress Index v1 | Artificial Analysis | 296.1 | 317.4 |
| Vals RSI Index | Vals AI | – | 18.3% |
| Vals Science Index | Vals AI | 27.4%† | – |
| Vals Multimodal Index | Vals AI | 56.1%† | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 29.8% | – |
| MASK | Scale AI SEAL | 42.4% | – |
Data as of 2026-11-10. Best configuration of each model; every score links to its source on the model pages.
What changed
- Gemini 3.7 Flash (high): BenchLeader Index 62.6 → 62.1.
- Gemini 3.7 Flash (high): 1 score revision.
- Gemini 3.1 Pro: 1 new result (Arena Search).
- Gemini 3.7 Flash (high): 1 new result (Arena Document).
- Gemini 3.1 Pro: 1 score revision.
- Gemini 3.1 Pro: 2 score revisions.
- Gemini 3.7 Flash (high): 1 new result (MirrorCode).
- Gemini 3.1 Pro: BenchLeader Index 61.6 → 60.5.
- Gemini 3.1 Pro: 2 new results (AA Progress Index v1, HLE-Diamond (AA)).
- Gemini 3.7 Flash (high): 1 new result (AA Progress Index v1).
Related comparisons
- Gemini 3.1 Pro vs Claude Fable 5.2
- Gemini 3.1 Pro vs Claude Opus 5.5
- Gemini 3.1 Pro vs Gemini 4 Carbon
- Gemini 3.7 Flash vs Claude Fable 5.2
- Gemini 3.7 Flash vs Claude Opus 5.5
- Gemini 3.7 Flash vs Gemini 4 Carbon
- Gemini 3.7 Flash vs GPT-6 Astra
- Gemini 3.1 Pro vs GPT-6.1 Sol
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Gemini 3.1 Pro vs Gemini 3.7 Flash: questions
- Is Gemini 3.1 Pro better than Gemini 3.7 Flash?
- Gemini 3.1 Pro and Gemini 3.7 Flash (high) are neck and neck on quality (60.5 vs 62.1, within the ±5.5 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 10 Nov 2026, so check the category scores for your use.
- Gemini 3.1 Pro vs Gemini 3.7 Flash: which is cheaper, and will that change?
- Gemini 3.7 Flash is cheaper today: $1.50 against $4.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Gemini 3.1 Pro or Gemini 3.7 Flash better for coding?
- Gemini 3.7 Flash scores higher in coding (63 vs 57 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1446 for Gemini 3.1 Pro against 1588 for Gemini 3.7 Flash.
- Which is faster, Gemini 3.1 Pro or Gemini 3.7 Flash?
- Gemini 3.1 Pro streams faster (108 vs 86 output tokens per second, Artificial Analysis 7-day median).
- Is Gemini 3.1 Pro or Gemini 3.7 Flash better for agents and tool use?
- Gemini 3.1 Pro scores higher in agentic tasks (57 vs 48 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 80.2% for Gemini 3.1 Pro against 12.4% for Gemini 3.7 Flash.
More questions (4)
- Which is better for long documents?
- They are level on long-context benchmarks (66 vs 66). Both accept 1,048,576 tokens of context.
- Which has lower latency?
- Gemini 3.7 Flash answers first: 2.5 s against 13.7 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Gemini 3.1 Pro vs Gemini 3.7 Flash?
- Across the 36 benchmarks both have, Gemini 3.1 Pro leads 7, Gemini 3.7 Flash leads 29. Every overall index that rates both puts Gemini 3.7 Flash ahead.
- When was Gemini 3.7 Flash released?
- Gemini 3.7 Flash was released on 13 Aug 2026.