BenchLeader
Menu

Gemini 3.8 Flash vs Gemini 4 Hydrogen

Full profiles: Gemini 3.8 Flash (Google) and Gemini 4 Hydrogen (Google): every benchmark, price and rank.

Updated · Data from 12 sources

As of 13 Nov 2026, Gemini 3.8 Flash and Gemini 4 Hydrogen are neck and neck on quality (61.7 vs 63.4 on the BenchLeader Index, inside the ±4.1 margin), and Gemini 3.8 Flash is 2.0× cheaper at $1.50 vs $3.00 per 1M tokens blended. Gemini 4 Hydrogen leads in every category we measure both in. Gemini 3.8 Flash's current price runs until 31 Dec 2026, then $3.00.

Key facts (as of 13 Nov 2026)

  • Gemini 3.8 Flash scores 61.7 on the BenchLeader Index and Gemini 4 Hydrogen scores 63.4 (BenchLeader, 13 Nov 2026).
  • Gemini 3.8 Flash costs $1.50 per 1M tokens blended until 31 Dec 2026, then $3.00 (https://ai.google.dev/gemini-api/docs/pricing, checked 20 Oct 2026).
  • Gemini 4 Hydrogen costs $3.00 per 1M tokens blended (list price, 13 Nov 2026).
  • Gemini 4 Hydrogen streams ~1.5× faster than Gemini 3.8 Flash (Artificial Analysis, 7-day median, 13 Nov 2026).
  • Of the 20 benchmarks both models have, Gemini 4 Hydrogen leads 15 (BenchLeader, 13 Nov 2026).
  • On Arena WebDev, Gemini 3.8 Flash is rated 1584 and Gemini 4 Hydrogen 1704 (LMArena, as published by 13 Nov 2026).
Gemini 3.8 Flash benchmarks high Google Gemini 4 Hydrogen benchmarks high GoogleAdd a third model

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 casePickWhyConfidence
CodingGemini 4 Hydrogen69 vs 60 category score; Arena WebDev 1704 vs 1584High
Agents & tool useGemini 4 Hydrogen62 vs 51 category score; Terminal-Bench 56.4% vs 18.2%High
MathsGemini 4 Hydrogen68 vs 62 category score; Arena Maths 1495 vs 1519Medium
ReasoningGemini 4 Hydrogen70 vs 63 category score; HLE-Diamond (AA) 54.9% vs 33.1%Medium
KnowledgeGemini 4 Hydrogen69 vs 64 category score; AA-Omniscience v1.1 43.7 vs 23.3Medium
Chat quality (human preference)Either69 vs 68 on Arena votesEither
Lowest costGemini 3.8 Flash (until 31 Dec 2026)$1.50 vs $3.00 per 1M blended; agentic coding session $0.060 vs $0.120; from 1 Jan 2027: $3.00 vs $3.00–
Fastest first answerEither17.0 s (Gemini 3.8 Flash (high)) vs 16.4 s (Gemini 4 Hydrogen (high)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Gemini 3.8 Flash (high) and Gemini 4 Hydrogen (high) are neck and neck on quality (61.7 vs 63.4, within the ±4.1 margin).
  • Gemini 3.8 Flash (high) is stronger in human preference.
  • Gemini 4 Hydrogen (high) is stronger in agents & tools, coding, knowledge, maths, reasoning.
  • Gemini 3.8 Flash (high) is 2.0× cheaper ($1.50 vs $3.00 per 1M blended) until 31 Dec 2026, then $3.00 vs $3.00 from 1 Jan 2027.
  • Gemini 4 Hydrogen (high) streams 1.6× faster (180 vs 111 tokens per second).

At a glance

SpecGemini 3.8 FlashGemini 4 Hydrogen
DeveloperGoogleGoogle
Released2 Sep 202612 Nov 2026
Context window (tokens)1,048,5761,048,576
Compared atGemini 3.8 Flash (high)Gemini 4 Hydrogen (high)
Open weightsNoNo
Price in / out / cached input per 1M$0.750 / $3.75 / $0.075$1.50 / $7.50 / $0.075
Announced price change$1.50 / $7.50 from 1 Jan 2027None announced

Limits, availability and announced prices from the vendors' own pages: ai.google.dev pricing (checked 20 Oct 2026). Everything else is from the day's data.

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.

CategoryGemini 3.8 Flash (high)Gemini 4 Hydrogen (high)Leader
Reasoning62.969.6Gemini 4 Hydrogen
Coding60.068.8Gemini 4 Hydrogen
Agents & tools51.461.9Gemini 4 Hydrogen
Maths61.667.8Gemini 4 Hydrogen
Knowledge63.868.5Gemini 4 Hydrogen
Human preference69.068.2level
Multimodal67.5–not comparable
Long context65.3–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.

2550 avg75ReasoningCodingAgents & toolsMathsKnowledgeHumanpreferenceGemini 3.8 Flash, Reasoning: 62.9Gemini 3.8 Flash, Coding: 60.0Gemini 3.8 Flash, Agents & tools: 51.4Gemini 3.8 Flash, Maths: 61.6Gemini 3.8 Flash, Knowledge: 63.8Gemini 3.8 Flash, Human preference: 69.0Gemini 4 Hydrogen, Reasoning: 69.6Gemini 4 Hydrogen, Coding: 68.8Gemini 4 Hydrogen, Agents & tools: 61.9Gemini 4 Hydrogen, Maths: 67.8Gemini 4 Hydrogen, Knowledge: 68.5Gemini 4 Hydrogen, Human preference: 68.2

Not drawn, measured for fewer than two of these models: Instruction following, Long context, Multimodal.

Model, click to add or removeIndex

Where sources disagree

Across the 20 benchmarks both have, Gemini 3.8 Flash leads 5, Gemini 4 Hydrogen leads 15. Independent sources split: Gemini 3.8 Flash leads on Arena Text; Gemini 4 Hydrogen leads on AA Intelligence Index v5.0 and Vals Index.

Overall indexGemini 3.8 Flash (high)Gemini 4 Hydrogen (high)Leader
BenchLeader Index61.763.4neck and neck
AA Intelligence Index v5.030.950.8Gemini 4 Hydrogen
Vals Index56.068.5†Gemini 4 Hydrogen
Arena Text14971490Gemini 3.8 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.

MeasureGemini 3.8 Flash (high)Gemini 4 Hydrogen (high)
Blended price per 1M tokens, until 31 Dec 2026$1.50$3.00
Input / output per 1M$0.750 / $3.75$1.50 / $7.50
Blended price per 1M from 1 Jan 2027$3.00$3.00
Input / output per 1M from 1 Jan 2027$1.50 / $7.50$1.50 / $7.50
Cached input per 1M$0.075$0.075
Price per task: chat reply$0.0014$0.0029
Price per task: summarize a 30-page report$0.011$0.022
Price per task: code edit$0.010$0.020
Price per task: agentic coding session$0.060$0.120
Price per task: structured extraction$0.0023$0.0045
Agentic coding session, input mostly cached$0.030$0.056
Cost per benchmark task (AA Index)$7.77$1.87
Output tokens per task (AA Index)99k85k
Of which reasoning60k48k
Time per task (AA Index)15 min8 min
Output speed111 tok/s180 tok/s
Time to first answer17.0 s16.4 s
Time to a complete answer52.2 s63.3 s

Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $7.77 on Gemini 3.8 Flash (high) and $1.87 on Gemini 4 Hydrogen (high), and Gemini 3.8 Flash (high) wrote 1.2× as many output tokens for the same work.

Each model is compared at its best-ranked reasoning-effort setting. Gemini 3.8 Flash (high) scores 61.7 and takes 17 s to its first answer token; Gemini 3.8 Flash (medium) scores 60.2 in 12 s, Gemini 3.8 Flash (low) scores 55.8 in 4 s. Gemini 4 Hydrogen (high) scores 63.4 and takes 16 s to its first answer token. 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

151 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.

MetricPublisherGemini 3.8 Flash (high)Gemini 4 Hydrogen (high)
BenchLeader IndexBenchLeader61.763.4
Agents & tools scoreBenchLeader51.461.9
Coding scoreBenchLeader60.068.8
Human preference scoreBenchLeader69.068.2
Knowledge scoreBenchLeader63.868.5
Long context scoreBenchLeader65.3–
Maths scoreBenchLeader61.667.8
Multimodal scoreBenchLeader67.5–
Reasoning scoreBenchLeader62.969.6
Blended price $/MBenchLeader$1.50$3.00
Output speedBenchLeader111 tok/s180 tok/s
Time to first answerBenchLeader17.0 s16.4 s
Context windowBenchLeader1,048,576 tokens1,048,576 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub95.4%–
Humanity's Last ExamScale AI / CAIS44.5%†–
SimpleBenchSimpleBench82.4%†–
Arena Hard PromptsArena (LMArena)15181514
LiveBench 2.0 ReasoningLiveBench89.3%†88.6%†
LiveBench 1.0 ReasoningLiveBench89.3%–
GPQA Diamond (AA)Artificial Analysis95.3%–
Humanity's Last Exam (AA)Artificial Analysis47.8%–
HLE-Diamond (AA)Artificial Analysis33.1%54.9%
GPQA Diamond (Vals)Vals AI94.4%–
ARC-AGI-1ARC Prize98.5%–
ARC-AGI-2ARC Prize89.2%–
ARC-AGI-3ARC Prize35.0%–
CritPtArtificial Analysis18.3%–
CritPt v2 hard subset (AA)Artificial Analysis46.5%61.5%
MysteryMechanismVals AI36.5%–
Chess PuzzlesEpoch AI Benchmarking Hub58.0%–
Mystery Game PuzzlesEpoch AI Benchmarking Hub47.0%–
LMCAEpoch AI Benchmarking Hub51.5%†–
DTBenchEpoch AI Benchmarking Hub95.7%†–
Coding
SciCodeSciCode56.6%–
WeirdMLWeirdML84.8%–
FrontierCodeCognition41.2%†–
Arena CodingArena (LMArena)15311518
Arena WebDevArena (LMArena)15841704
Show all 151 benchmarks
MetricPublisherGemini 3.8 Flash (high)Gemini 4 Hydrogen (high)
Coding
LiveBench 2.0 CodingLiveBench73.3%†78.5%†
LiveBench 1.0 CodingLiveBench72.5%–
LiveCodeBenchVals AI89.5%–
IOIVals AI56.9%–
SWE-bench (Vals)Vals AI80.0%–
Code MigrationVals AI36.5%–
Code Migration v1.1Vals AI38.6%–
ProgramBenchVals AI1.0%–
SWE-bench MultimodalVals AI40.8%–
Vibe Code Bench 1-100Vals AI18.8%–
Vibe Code Bench v1.1Vals AI78.7%87.8%†
SWE Atlas: Codebase QnAScale AI SEAL47.0%†–
SWE Atlas: RefactoringScale AI SEAL44.8%†–
SWE Atlas: Test WritingScale AI SEAL53.7%†–
MirrorCodeEpoch AI Benchmarking Hub38.9%–
Surface Evolver BenchEpoch AI Benchmarking Hub76.9%–
DeepSWE v1.1Epoch AI Benchmarking Hub73.8%–
CursorBenchCursor39.6%–
ALE-BenchEpoch AI Benchmarking Hub1269.7–
FrontierSWEEpoch AI Benchmarking Hub19.6%–
Terminal-Bench 4.0 (AA)Artificial Analysis19.7%54.0%
Terminal-Bench 2.1 (AA)Artificial Analysis87.6%–
Startrise FrontendStartrise64.6%†–
GameBench 2 OverallDuelLab GameBench 258.7†–
GameBench 2 by reasoning settingDuelLab GameBench 263.7–
Design Arena WebsiteDesign Arena1291†1316†
FrontierCode 1.1 (AA)Artificial Analysis36.4%–
FrontierCode 2.0 (AA)Artificial Analysis18.7%36.3%
Agents & tools
Terminal-BenchTerminal-Bench18.2%56.4%
Remote Labor IndexScale AI / CAIS5.8%†–
APEX-AgentsMercor64.3%†–
Arena AgentArena (LMArena)2.3–
LiveBench Long AgenticLiveBench16.5%†61.8%†
LiveBench BrowserLiveBench45.1%†77.1%†
LiveBench Agentic CodingLiveBench54.2%–
Terminal-Bench 2.1 (Vals)Vals AI81.3%–
HiL-BenchScale AI SEAL41.5%†–
GDPval-AA v2.1, normalised scoreArtificial Analysis46.8%–
GDPval-AA v2.1Artificial Analysis1445–
τ³-Banking (AA)Artificial Analysis45.0%–
ITBench SRE (AA)Artificial Analysis52.5%–
Analyst Agent (AA)Artificial Analysis46.3%–
CUA-benchVals AI4.2%–
CyberBenchVals AI43.8%–
Finance Agent v2Vals AI61.4%–
Finance Agent v2.1Vals AI65.8%65.1%†
Harvey's Legal Agent Benchmark v1.1Vals AI12.1%11.3%†
Harvey's Legal Agent Benchmark v1Vals AI10.0%–
Legal Research BenchVals AI38.9%53.9%†
SkillsBenchVals AI58.0%–
SREBenchVals AI11.1%–
Tax Agent BenchVals AI32.1%52.3%†
Terminal-Bench 4.0 (Vals)Vals AI19.7%54.0%†
Terminal-Bench ScienceVals AI8.6%–
Time Horizon Index: KSPVals AI18.8%–
Vending-Bench 2Andon Labs5093.8†–
AutomationBenchArtificial Analysis59.9%74.4%
GDP.pdf (AA)Artificial Analysis21.0%30.0%
Harvey LAB-AA v1.1Harvey0.0%–
AA-Briefcase v1.1Artificial Analysis12031497
Design Arena AgenticDesign Arena1217†–
Terminal-Bench-Science 0.1 (AA)Artificial Analysis10.0%41.4%
CWE-Bench (AA)Artificial Analysis52.5%–
DeepSec-Bench (AA)Artificial Analysis14.5%–
CyberGym end-to-end (AA)Artificial Analysis8.4%–
FACTS SearchGoogle80.4%†–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub68.4%–
FrontierMath Tier 4Epoch AI Benchmarking Hub21.9%–
OTIS Mock AIMEEpoch AI Benchmarking Hub98.9%–
ProofBenchVals AI48.0%†–
LiveBench MathematicsLiveBench91.6%–
Arena MathsArena (LMArena)15191495
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub69.7%–
LiveBench 2.0 Data AnalysisLiveBench54.0%†83.2%†
LiveBench ResearchLiveBench79.4%†86.8%†
LiveBench 1.0 Data AnalysisLiveBench54.0%–
AA-Omniscience v1.1Artificial Analysis23.343.7
MMLU-ProVals AI90.2%–
LegalBenchVals AI87.0%–
TaxEvalVals AI74.5%–
PRBench FinanceScale AI SEAL49.5%†–
PRBench LegalScale AI SEAL48.9%†–
BioMysteryBenchVals AI62.2%–
Excel Modeling BenchmarkVals AI72.2%–
Excel Modeling Benchmark v1.1Vals AI72.9%75.8%†
MedCodeVals AI48.1%–
MedScribeVals AI84.5%–
Public Benefits Bench v1.1Vals AI65.3%–
GDP.pdfEpoch AI Benchmarking Hub23.2%–
AA-Omniscience v1.1: accuracyArtificial Analysis50.5%48.2%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis44.9%91.3%
HealthBench Professional (AA)OpenAI51.7%–
Instruction following
LiveBench LanguageLiveBench87.8%–
LiveBench Instruction FollowingLiveBench81.4%–
Arena Instruction FollowingArena (LMArena)14921503
Human preference
Arena TextArena (LMArena)14971490
Arena Creative WritingArena (LMArena)14881492
Arena Multi-turnArena (LMArena)15051476
Arena Longer QueriesArena (LMArena)15111514
Multimodal
Arena VisionArena (LMArena)1317–
LiveBench DesignLiveBench69.7%†77.3%†
MMMU-ProArtificial Analysis85.6%–
MMMU-Pro (Vals)Vals AI89.1%–
MortgageTaxVals AI65.3%–
SAGEVals AI88.2%–
AudioMCScale AI SEAL60.4%–
AudioMC (text output)Scale AI SEAL60.4%–
GeoBenchGeoBench86.0%†–
Blueprint-Bench 2Epoch AI Benchmarking Hub38.6%†–
Code2Video-AAArtificial Analysis899–
AA-3DArtificial Analysis974–
Long context
AA-LCR v1.1Artificial Analysis81.3%–
Arena DocumentArena (LMArena)1469–
MLCRArtificial Analysis10.6%21.1%†
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub158.8†–
LiveBench 2.0LiveBench57.3%†77.0%†
LiveBench 1.0LiveBench75.8%–
AA Intelligence Index v5.0Artificial Analysis30.950.8
AA Intelligence Index v4.3.2Artificial Analysis40.9–
AA Progress Index v1Artificial Analysis320.3–
Vals IndexVals AI56.068.5†
Vals Finance IndexVals AI60.5%66.1%†
Vals RSI IndexVals AI19.9%–
Vals Science IndexVals AI35.8%–
Vals Router IndexVals AI56.0%†–
AA Cyber IndexArtificial Analysis25.1–

Data as of 2026-11-13. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261110-2acac640a228 (https://benchmarks.duellab.org/releases/gb2-20261110-2acac640a228/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • Gemini 3.8 Flash: list price moves to $1.50 / $7.50 per 1M input / output tokens (announced).
  • Gemini 4 Hydrogen (high): 10 new results (Arena Text, Arena Hard Prompts, Arena Coding…).
  • Gemini 3.8 Flash (high): 1 new result (MirrorCode).
  • Gemini 3.8 Flash (high): 2 score revisions.
  • Gemini 3.8 Flash (high): 1 new result (Arena Document); 1 score revision.
  • Gemini 3.8 Flash (high): 2 score revisions.
  • Gemini 3.8 Flash (high): 1 new result (MirrorCode).
  • Gemini 3.8 Flash (high): 1 score revision.
  • Gemini 3.8 Flash (high): BenchLeader Index 63.3 → 61.8.
  • Gemini 3.8 Flash (high): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; cost per task moved from $3.49/M to $7.77/M (123% up).

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

Gemini 3.8 Flash vs Gemini 4 Hydrogen: questions

Is Gemini 3.8 Flash better than Gemini 4 Hydrogen?
Gemini 3.8 Flash (high) and Gemini 4 Hydrogen (high) are neck and neck on quality (61.7 vs 63.4, within the ±4.1 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 13 Nov 2026, so check the category scores for your use.
Gemini 3.8 Flash vs Gemini 4 Hydrogen: which is cheaper, and will that change?
Gemini 3.8 Flash is cheaper today: $1.50 against $3.00 per million tokens, blended at three input tokens per output token. Gemini 3.8 Flash's current price runs until 31 Dec 2026, then $3.00.
Is Gemini 3.8 Flash or Gemini 4 Hydrogen better for coding?
Gemini 4 Hydrogen scores higher in coding (69 vs 60 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1584 for Gemini 3.8 Flash against 1704 for Gemini 4 Hydrogen.
Which is faster, Gemini 3.8 Flash or Gemini 4 Hydrogen?
Gemini 4 Hydrogen streams ~1.5× faster (180 vs 111 output tokens per second, Artificial Analysis 7-day median).
Is Gemini 3.8 Flash or Gemini 4 Hydrogen better for agents and tool use?
Gemini 4 Hydrogen scores higher in agentic tasks (62 vs 51 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 18.2% for Gemini 3.8 Flash against 56.4% for Gemini 4 Hydrogen.
More questions (4)
Which is better for long documents?
Both accept 1,048,576 tokens of context.
Which has lower latency?
Gemini 4 Hydrogen answers first: 16.4 s against 17.0 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 3.8 Flash vs Gemini 4 Hydrogen?
Across the 20 benchmarks both have, Gemini 3.8 Flash leads 5, Gemini 4 Hydrogen leads 15. Independent sources split: Gemini 3.8 Flash leads on Arena Text; Gemini 4 Hydrogen leads on AA Intelligence Index v5.0 and Vals Index.
When was Gemini 4 Hydrogen released?
Gemini 4 Hydrogen was released on 12 Nov 2026.