Gemini 3.7 Flash vs Gemini 4 Neon
Full profiles: Gemini 3.7 Flash (Google) and Gemini 4 Neon (Google): every benchmark, price and rank.
Updated · Data from 10 sources
As of 19 Nov 2026, Gemini 4 Neon is ahead of Gemini 3.7 Flash on quality (67.3 vs 61.9 on the BenchLeader Index), but Gemini 3.7 Flash is 4.0× cheaper at $1.50 vs $6.00 per 1M tokens blended. Gemini 4 Neon leads in every category we measure both in.
Key facts (as of 19 Nov 2026)
- Gemini 3.7 Flash scores 61.9 on the BenchLeader Index and Gemini 4 Neon scores 67.3 (BenchLeader, 19 Nov 2026).
- Gemini 3.7 Flash costs $1.50 per 1M tokens blended (list price, 19 Nov 2026).
- Gemini 4 Neon costs $6.00 per 1M tokens blended (list price, 19 Nov 2026).
- Of the 15 benchmarks both models have, Gemini 4 Neon leads 15 (BenchLeader, 19 Nov 2026).
- On Arena WebDev, Gemini 3.7 Flash is rated 1588 and Gemini 4 Neon 1764 (LMArena, as published by 19 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 4 Neon | 73 vs 62 category score; Arena WebDev 1764 vs 1588 | High |
| Agents & tool use | Gemini 4 Neon | 76 vs 49 category score; Arena Agent 18.4 vs -3.1 | High |
| Reasoning | Gemini 4 Neon | 72 vs 67 category score; Arena Hard Prompts 1546 vs 1510 | Medium |
| Knowledge | Gemini 4 Neon | 72 vs 66 category score | Medium |
| Chat quality (human preference) | Gemini 4 Neon | 72 vs 68 on Arena votes | Medium |
| Lowest cost | Gemini 3.7 Flash | $1.50 vs $6.00 per 1M blended; agentic coding session $0.060 vs $0.240 | – |
| Fastest first answer | Gemini 3.7 Flash | 2.1 s (Gemini 3.7 Flash (high)) vs 40.6 s (Gemini 4 Neon (high)) 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 4 Neon (high) leads on quality: 67.3 vs 61.9.
- Gemini 4 Neon (high) is stronger in agents & tools, coding, human preference, knowledge, reasoning.
- Gemini 3.7 Flash (high) is 4.0× cheaper ($1.50 vs $6.00 per 1M blended).
At a glance
| Spec | Gemini 3.7 Flash | Gemini 4 Neon |
|---|---|---|
| Developer | ||
| Released | 13 Aug 2026 | 18 Nov 2026 |
| Context window (tokens) | 1,048,576 | 1,048,576 |
| Compared at | Gemini 3.7 Flash (high) | Gemini 4 Neon (high) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $0.750 / $3.75 / $0.075 | $3.00 / $15.00 / $0.150 |
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 3.7 Flash (high) | Gemini 4 Neon (high) | Leader |
|---|---|---|---|
| Reasoning | 67.3 | 71.9 | Gemini 4 Neon |
| Coding | 62.1 | 73.4 | Gemini 4 Neon |
| Agents & tools | 48.6 | 76.2 | Gemini 4 Neon |
| Maths | 62.3 | – | not comparable |
| Knowledge | 65.6 | 72.3 | Gemini 4 Neon |
| Human preference | 68.2 | 71.7 | Gemini 4 Neon |
| Multimodal | 67.0 | – | not comparable |
| Long context | 65.5 | – | not comparable |
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: Maths, Instruction following, Long context, Multimodal.
Where sources disagree
Across the 15 benchmarks both have, Gemini 3.7 Flash leads 0, Gemini 4 Neon leads 15. Every overall index that rates both puts Gemini 4 Neon ahead.
| Overall index | Gemini 3.7 Flash (high) | Gemini 4 Neon (high) | Leader |
|---|---|---|---|
| BenchLeader Index | 61.9 | 67.3 | Gemini 4 Neon |
| Arena Text | 1490 | 1519 | Gemini 4 Neon |
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 3.7 Flash (high) | Gemini 4 Neon (high) |
|---|---|---|
| Blended price per 1M tokens | $1.50 | $6.00 |
| Input / output per 1M | $0.750 / $3.75 | $3.00 / $15.00 |
| Cached input per 1M | $0.075 | $0.150 |
| Price per task: chat reply | $0.0014 | $0.0057 |
| Price per task: summarize a 30-page report | $0.011 | $0.045 |
| Price per task: code edit | $0.010 | $0.041 |
| Price per task: agentic coding session | $0.060 | $0.240 |
| Price per task: structured extraction | $0.0023 | $0.0090 |
| Agentic coding session, input mostly cached | $0.030 | $0.112 |
| Cost per benchmark task (AA Index) | – | $3.74 |
| Output tokens per task (AA Index) | – | 90k |
| Of which reasoning | – | 46k |
| Time per task (AA Index) | – | 15 min |
| Output speed | 96 tok/s | 101 tok/s |
| Time to first answer | 2.1 s | 40.6 s |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.7 Flash (high) scores 61.9 and takes 2 s to its first answer token; Gemini 3.7 Flash (medium) scores 60.0 in 2 s, Gemini 3.7 Flash (low) scores 57.1 in 2 s. Gemini 4 Neon (high) scores 67.3 and takes 41 s to its first answer token; Gemini 4 Neon (xhigh) scores 64.3 in 57 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
111 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 3.7 Flash (high) | Gemini 4 Neon (high) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 61.9 | 67.3 |
| Agents & tools score | BenchLeader | 48.6 | 76.2 |
| Coding score | BenchLeader | 62.1 | 73.4 |
| Human preference score | BenchLeader | 68.2 | 71.7 |
| Knowledge score | BenchLeader | 65.6 | 72.3 |
| Long context score | BenchLeader | 65.5 | – |
| Maths score | BenchLeader | 62.3 | – |
| Multimodal score | BenchLeader | 67.0 | – |
| Reasoning score | BenchLeader | 67.3 | 71.9 |
| Blended price $/M | BenchLeader | $1.50 | $6.00 |
| Output speed | BenchLeader | 96 tok/s | 101 tok/s |
| Time to first answer | BenchLeader | 2.1 s | 40.6 s |
| Context window | BenchLeader | 1,048,576 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 94.8% | – |
| Arena Hard Prompts | Arena (LMArena) | 1510 | 1546 |
| LiveBench 2.0 Reasoning | LiveBench | – | 89.6%† |
| LiveBench 1.0 Reasoning | LiveBench | 87.8% | – |
| GPQA Diamond (AA) | Artificial Analysis | 94.5% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 47.9% | – |
| HLE-Diamond (AA) | Artificial Analysis | – | 68.4% |
| GPQA Diamond (Vals) | Vals AI | 93.9% | – |
| ARC-AGI-1 | ARC Prize | 95.5% | – |
| ARC-AGI-2 | ARC Prize | 84.6% | – |
| CritPt | Artificial Analysis | 14.3% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | – | 59.0% |
| Arena Expert | Arena (LMArena) | 1519 | 1541 |
| Chess Puzzles | Epoch AI Benchmarking Hub | 66.0% | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 37.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 50.4% | – |
| DTBench | Epoch AI Benchmarking Hub | 96.8% | – |
| ForecastBench | ForecastBench | 65.7% | – |
| Coding | |||
| SciCode | SciCode | 57.2% | – |
| FrontierCode | Cognition | 43.6%† | – |
| Arena Coding | Arena (LMArena) | 1521 | 1558 |
| Arena WebDev | Arena (LMArena) | 1588 | 1764 |
| LiveBench 2.0 Coding | LiveBench | – | 89.7%† |
| LiveBench 1.0 Coding | LiveBench | 78.9% | – |
| LiveCodeBench | Vals AI | 88.7% | – |
Show all 111 benchmarks
| Metric | Publisher | Gemini 3.7 Flash (high) | Gemini 4 Neon (high) |
|---|---|---|---|
| Coding | |||
| IOI | Vals AI | 67.8% | – |
| SWE-bench (Vals) | Vals AI | 80.8% | – |
| Code Migration | Vals AI | 34.8% | – |
| ProgramBench | Vals AI | 0.0% | – |
| Vibe Code Bench v1.1 | Vals AI | 70.4% | – |
| MirrorCode | Epoch AI Benchmarking Hub | 41.1% | – |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 65.3% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 904.3 | – |
| FrontierSWE | Epoch AI Benchmarking Hub | 20.3% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 13.6% | 61.6% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 85.8% | – |
| Design Arena Website | Design Arena | 1293† | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | – | 43.3% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 12.4% | – |
| Remote Labor Index | Scale AI / CAIS | 5.0%† | – |
| APEX-Agents | Mercor | 67.8%† | – |
| Arena Agent | Arena (LMArena) | -3.1 | 18.4 |
| LiveBench Long Agentic | LiveBench | – | 82.6%† |
| LiveBench Browser | LiveBench | – | 95.6%† |
| LiveBench Agentic Coding | LiveBench | 58.3% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 77.5% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1404 | – |
| τ³-Banking (AA) | Artificial Analysis | 32.8% | – |
| Analyst Agent (AA) | Artificial Analysis | 60.0% | – |
| CyberBench | Vals AI | 43.8% | – |
| Finance Agent v2 | Vals AI | 59.0% | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 8.8% | – |
| Legal Research Bench | Vals AI | 34.6% | – |
| SkillsBench | Vals AI | 65.9% | – |
| SREBench | Vals AI | 4.6% | – |
| Tax Agent Bench | Vals AI | 22.8% | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | 12.1% | – |
| Terminal-Bench Science | Vals AI | 7.1% | – |
| Arena Agent Task Completion | Arena (LMArena) | -3.4 | 19.2 |
| AutomationBench | Artificial Analysis | 62.0% | 79.2% |
| GDP.pdf (AA) | Artificial Analysis | 23.6% | 33.4% |
| EnterpriseOps-Gym | Artificial Analysis | 50.4%† | – |
| AA-Briefcase v1.1 | Artificial Analysis | 1115 | 1632 |
| Design Arena Agentic | Design Arena | 1213† | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | – | 65.7% |
| MKA (AA) | Artificial Analysis | – | 54.6% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 71.6% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 36.6% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 97.2% | – |
| ProofBench | Vals AI | 58.0%† | – |
| LiveBench Mathematics | LiveBench | 93.5% | – |
| Arena Maths | Arena (LMArena) | 1502 | – |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 69.2% | – |
| LiveBench 2.0 Data Analysis | LiveBench | – | 85.3%† |
| LiveBench Research | LiveBench | – | 94.5%† |
| LiveBench 1.0 Data Analysis | LiveBench | 68.0% | – |
| AA-Omniscience v1.1 | Artificial Analysis | – | 53.8 |
| MMLU-Pro | Vals AI | 90.1% | – |
| LegalBench | Vals AI | 87.3% | – |
| TaxEval | Vals AI | 74.7% | – |
| Excel Modeling Benchmark | Vals AI | 71.3% | – |
| MedCode | Vals AI | 53.4% | – |
| MedScribe | Vals AI | 83.9% | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | – | 59.4% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | – | 86.1% |
| Instruction following | |||
| LiveBench Language | LiveBench | 85.5% | – |
| LiveBench Instruction Following | LiveBench | 79.9% | – |
| Arena Instruction Following | Arena (LMArena) | 1486 | 1522 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1490 | 1519 |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1720† | – |
| Arena Creative Writing | Arena (LMArena) | 1487 | 1514 |
| Arena Multi-turn | Arena (LMArena) | 1490 | 1529 |
| Arena Longer Queries | Arena (LMArena) | 1502 | 1521 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1313 | – |
| LiveBench Design | LiveBench | – | 81.5%† |
| MMMU-Pro | Artificial Analysis | 85.5% | – |
| MMMU-Pro (Vals) | Vals AI | 89.0% | – |
| MortgageTax | Vals AI | 66.7% | – |
| Humanity's Sixth Sense | Scale AI SEAL | 39.8% | 75.9%† |
| Furniture Assembly | Epoch AI Benchmarking Hub | 26.7% | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 81.7% | – |
| Arena Document | Arena (LMArena) | 1468 | – |
| MLCR | Artificial Analysis | 12.2% | 28.9% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 158.4† | – |
| LiveBench 2.0 | LiveBench | – | 87.4%† |
| LiveBench 1.0 | LiveBench | 78.8% | – |
| AA Intelligence Index v5.0.1 | Artificial Analysis | – | 59.6 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 39.1 | – |
| AA Progress Index v1 | Artificial Analysis | 317.4 | – |
| Vals RSI Index | Vals AI | 18.3% | – |
| Safety & honesty | |||
| DistressBench | Scale AI SEAL | 69.0%† | – |
Data as of 2026-11-19. Best configuration of each model; every score links to its source on the model pages.
What changed
- Gemini 4 Neon (high): 13 new results (MLCR, MKA (AA), Arena Text…).
- Gemini 4 Neon (high): first results.
- Gemini 3.7 Flash (high): 1 score revision.
- Gemini 3.7 Flash (high): 1 score revision.
- Gemini 3.7 Flash (high): 2 new results (GDPval-AA v2.1, AA-Briefcase v1.1).
- Gemini 3.7 Flash (high): BenchLeader Index 62.6 → 62.1.
- Gemini 3.7 Flash (high): 1 score revision.
- Gemini 3.7 Flash (high): 1 new result (Arena Document).
- Gemini 3.7 Flash (high): 1 new result (MirrorCode).
- Gemini 3.7 Flash (high): 1 new result (AA Progress Index v1).
Related comparisons
- Gemini 3.7 Flash vs Claude Fable 5.2
- Gemini 3.7 Flash vs GPT-6.2 Astra
- Gemini 3.7 Flash vs Gemini 4 Carbon
- Gemini 4 Neon vs Claude Fable 5.2
- Gemini 4 Neon vs GPT-6.2 Astra
- Gemini 4 Neon vs Gemini 4 Carbon
- Gemini 4 Neon vs Claude Opus 5.5
- Gemini 3.7 Flash vs Gemini 4 Hydrogen
Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs
Gemini 3.7 Flash vs Gemini 4 Neon: questions
- Is Gemini 4 Neon better than Gemini 3.7 Flash?
- Gemini 4 Neon (high) leads on quality: 67.3 vs 61.9. The BenchLeader Index combines every independent quality benchmark; Gemini 4 Neon (high) is ahead overall as of 19 Nov 2026, but check the category scores for your use.
- Gemini 3.7 Flash vs Gemini 4 Neon: which is cheaper, and will that change?
- Gemini 3.7 Flash is cheaper today: $1.50 against $6.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Gemini 3.7 Flash or Gemini 4 Neon better for coding?
- Gemini 4 Neon scores higher in coding (73 vs 62 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1588 for Gemini 3.7 Flash against 1764 for Gemini 4 Neon.
- Which is faster, Gemini 3.7 Flash or Gemini 4 Neon?
- Gemini 4 Neon streams faster (101 vs 96 output tokens per second, Artificial Analysis 7-day median).
- Is Gemini 3.7 Flash or Gemini 4 Neon better for agents and tool use?
- Gemini 4 Neon scores higher in agentic tasks (76 vs 49 on the category index, where 50 is average). On Arena Agent, the top shared agents & tools benchmark, it is -3.1 for Gemini 3.7 Flash against 18.4 for Gemini 4 Neon.
More questions (4)
- Which is better for long documents?
- Both accept 1,048,576 tokens of context.
- Which has lower latency?
- Gemini 3.7 Flash answers first: 2.1 s against 40.6 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 3.7 Flash vs Gemini 4 Neon?
- Across the 15 benchmarks both have, Gemini 3.7 Flash leads 0, Gemini 4 Neon leads 15. Every overall index that rates both puts Gemini 4 Neon ahead.
- When was Gemini 4 Neon released?
- Gemini 4 Neon was released on 18 Nov 2026.