Skip to main content
BenchLM
Data

Data as of November 21, 2026 · How the score is built

Superseded.Google has newer models in this line:Gemini 4 Hydrogen

Decision readingGemini 3.8 Flash scores 67.9 out of 100 and ranks #28 of 216. This profile shows 59 source-displayable benchmark rows; its strongest eligible category is Multilingual at #13. API pricing is $0.75 input and $3.75 output per million tokens, with cached input at $0.075.

Email me when its price, rank or availability changes.

Capability

67.9/100

field median 47.5#28 of 216 ranked models

Public

#28of 216

Verified #27 of 147

Price

$0.75input / $3.75 output

input median $0.60cached $0.075 · blended $2.25

Speed

115tok/s

field median 93.5 tok/sFirst token 19.76 s

Context

1Mtokens

field median 256,000Maximum output length is tracked separately

Decision snapshot

Where this model sits in the ranked field, its strongest published evidence, and what to validate before choosing it. The marker compares it with the current ranked catalog, not with an absolute quality threshold.

Strongest published evidence

Multilingual ranks #13. A well-rounded choice across a range of tasks.

Validate before choosing

59 published rows leave some tracked benchmark slots empty. Coding is its lowest eligible category at #36.

Source-linked · 59 displayable benchmark rows

Is Gemini 3.8 Flash right for your job?

Its place on the LLM Selector’s job rankings, the same order the job pages publish. Describe your own job to see where it stands for that.

Gemini 3.8 Flash has a newer version, Gemini 4 Hydrogen. Where Gemini 4 Hydrogen ranks by benchmark score:

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

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 scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #31 of 182Percentile 83rdWeight 22%11 benchmarksVerified
66.0
CodingRank #36 of 180Percentile 80thWeight 20%8 benchmarksVerified
65.5
ReasoningRank #24 of 161Percentile 86thWeight 17%4 benchmarksVerified
69.3
MultimodalRank #23 of 119Percentile 81stWeight 12%5 benchmarksVerified
66.8
KnowledgeRank #16 of 158Percentile 90thWeight 12%6 benchmarksVerified
71.5
MultilingualRank #13 of 88Percentile 86thWeight 7%Not listedUnlisted sources
66.2
Inst. FollowingRank #21 of 143Percentile 86thWeight 5%1 benchmarkVerified
65.6
MathRank #24 of 125Percentile 81stWeight 5%2 benchmarksVerified
64.3

59 of 952 tracked benchmark slots have displayable evidence · bars run 0–100

Coverage details

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.

  1. Agentic11/11 verified
  2. Coding8/8 verified
  3. Reasoning4/4 verified
  4. Multimodal5/5 verified
  5. Knowledge6/6 verified
  6. MultilingualNot measured
  7. Inst. Following1/1 verified
  8. Math2/2 verified
Verified sourceProvisionalNot measured

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 3.8 Flash category percentile values

  • Agentic83rd percentile
  • Coding80th percentile
  • Reasoning86th percentile
  • Multimodal81st percentile
  • Knowledge90th percentile
  • Multilingual86th percentile
  • Instruction following86th percentile
  • Math81st percentile

The dashed outline is median of 6 nearest peers.

Eligible category ranks

  1. Agentic#31/182
  2. Coding#36/180
  3. Reasoning#24/161
  4. Multimodal#23/119
  5. Knowledge#16/158
  6. Multilingual#13/88
  7. Inst. Following#21/143
  8. Math#24/125
Top decileTop quartileMid-fieldNot eligible

Benchmark ledger

Every result says who ran the test. For this model, Google ran 12 of 59. Coding opens by default. The tick marks the best source-verified result in the catalog; provisional leaders do not set the reference.

Overall rank
#28
of 216 on the leaderboard
Reported by Google
12
Shown with that label
Run by a benchmark
42
The benchmark’s own run
Independent runs
5
An outside evaluator’s run
Coding8 rows
Coding benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
DeepSWE73.8%

Best verified: Muse Wonder Full · 79.0%

Gap5.2 behindRun by DeepSWECounts5% ref. weight
Bug Hunt Bench18.0 fixes

Best verified: Gemini 4 Neon · 63.0 fixes

Gap45 behindRun by Bug Hunt BenchCounts5% ref. weight
FrontierSWE v219.6%

Best verified: Claude Fable 5.2 · 69.9%

Gap50.3 behindRun by FrontierSWE v2Counts4% ref. weight
Terminal-Bench 2.189.4%

Best verified: SWE-2 · 92.8%

Gap3.4 behindReported by GoogleCountsScored in Agentic
ScarfBenchScarfBench: Cross-Framework Application Migration in Enterprise Java66.0%

Best verified: Gemini 4 Argon · 75.2%

Gap9.2 behindRun by ScarfBenchShown only
CursorBench 3.269.2%

Best verified: Claude Fable 5.1 · 73.4%

Gap4.2 behindRun by CursorBench 3.2Shown only
LiveCodeBench (Vals)LiveCodeBench, Vals AI run89.5%

Best verified: Claude Fable 5.1 · 90.5%

Gap1 behindIndependent run: Vals AIShown only
SWE-bench (Vals)SWE-bench, Vals AI run80.0%

Best verified: Claude Opus 5 · 97.0%

Gap17 behindIndependent run: Vals AIShown only
Agentic11 rows
Agentic benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
Terminal-Bench 4.118.20%

Best verified: Claude Fable 5.2 · 65.80%

Gap47.6 behindRun by Terminal-Bench 4.1Counts4% ref. weight
Gert LabsGert Labs Composite Game Benchmark66.98%

Best verified: GPT-6 Astra · 91.96%

Gap25 behindRun by Gert LabsCounts3% ref. weight
Terminal-Bench 4.019.10%

Best verified: Claude Sonnet 5.5 · 70.60%

Gap51.5 behindReported by GoogleCounts3% ref. weight
Terminal-Bench 2.189.4%

Best verified: SWE-2 · 92.8%

Gap3.4 behindReported by GoogleCounts2% ref. weight
Finance Agent v261.4%

Best verified: Gemini 4 Argon · 65.4%

Gap4 behindReported by GoogleShown only
OSWorld 2.059.0%

Best verified: Kimi K3.1 Pro · 79.6%

Gap20.6 behindReported by GoogleShown only
Terminal-Bench 2.1 (Vals)Terminal-Bench 2.1, Vals AI run81.3%

Best verified: Claude Opus 5.5 · 87.6%

Gap6.3 behindIndependent run: Vals AIShown only
ApprenticeBenchApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job24%

Best verified: Claude Fable 5.1 · 72%

Gap48 behindRun by ApprenticeBenchShown only
Hermes IndexHermes Index: agentic capability and cost per task inside Hermes Agent38.1%

Best verified: Claude Opus 5.5 · 63.3%

Gap25.2 behindRun by Hermes IndexShown only
Hermes BenchHermes Bench: 150 agent tasks graded on the files and state the agent leaves behind62.2%

Best verified: Claude Opus 5.5 · 76.7%

Gap14.5 behindRun by Hermes BenchShown only
MazeBenchMazeBench: visual spatial reasoning in a 3D open world4%

Best verified: GPT-6 Astra · 24%

Gap20 behindRun by MazeBenchShown only
Reasoning4 rows
Reasoning benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
ARC-AGI-2Abstraction and Reasoning Corpus for AGI v289.2%

Best verified: GPT-6.2 Astra · 95.8%

Gap6.7 behindRun by ARC-AGI-2Counts2% ref. weight
ARC-AGI-1ARC-AGI-1 Semi-Private Evaluation98.50%

Best verified: Claude Fable 5 · 98.50%

GapBest verifiedRun by ARC-AGI-1Shown only
ARC-AGI-3Abstraction and Reasoning Corpus for AGI v310.4%

Best verified: GPT-6.2 Astra · 70.3%

Gap59.9 behindRun by ARC-AGI-3Shown only
ARC-AGI-3 Provider AdapterARC-AGI-3 Semi-Private: Provider Adapter harness35.00%

Best verified: GPT-6.2 Astra · 99.80%

Gap64.8 behindRun by ARC-AGI-3 Provider AdapterShown only
Multimodal5 rows
Multimodal benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
ParseBench-2ParseBench-2: severity-aware document parsing (LlamaIndex leaderboard)76.67%

Best verified: Claude Opus 5.5 · 87.11%

Gap10.4 behindRun by ParseBench-2Counts5% ref. weight
ParseBench-1ParseBench-1: document parsing for AI agents (LlamaIndex leaderboard, first release)71.20%

Best verified: Claude Opus 5.5 · 78.80%

Gap7.6 behindRun by ParseBench-1Counts5% ref. weight
CharXiv w/o toolsCharXiv Reasoning without tools86.2%

Best verified: Kimi K3.1 Pro · 89%

Gap2.8 behindReported by GoogleCounts3% ref. weight
ExtractBench (LlamaIndex)ExtractBench: schema-guided enterprise document extraction (LlamaIndex leaderboard)80.71%

Best verified: GPT-6.1 Sol · 95.31%

Gap14.6 behindRun by ExtractBench (LlamaIndex)Counts3% ref. weight
LVBench87.1%

Best verified: Gemini 4 Argon · 91.7%

Gap4.6 behindReported by GoogleCounts2% ref. weight
Knowledge6 rows
Knowledge benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
LABBench2LABBench2: An Improved Benchmark for AI Systems Performing Biology Research86.2%

Best verified: Gemini 4 Argon · 88.8%

Gap2.6 behindReported by GoogleCounts2% ref. weight
HLE-Verified54.9%

Best verified: Gemini 3.8 Flash · 54.9%

GapHighest listedReported by GoogleCounts2% ref. weight
BioMysteryBench (human-difficult)BioMysteryBench Human Difficult56.5%

Best verified: Gemini 3.8 Flash · 56.5%

GapHighest listedReported by GoogleCounts2% ref. weight
BioMysteryBench (human-solvable)BioMysteryBench Human Solvable88.8%

Best verified: Claude Opus 5 · 90.1%

Gap1.3 behindReported by GoogleShown only
GPQA Diamond (Vals)GPQA Diamond, Vals AI run94.4%

Best verified: Gemini 3.1 Pro · 95.5%

Gap1.1 behindIndependent run: Vals AIShown only
MMLU-Pro (Vals)MMLU-Pro, Vals AI run90.2%

Best verified: Claude Fable 5.1 · 92.4%

Gap2.2 behindIndependent run: Vals AIShown only
Inst. Following1 row
Inst. Following benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
Gray Swan IPI (15 attempts)Gray Swan Indirect Prompt Injection, attack success at 15 attempts5.5%

Best verified: Gemini 4 Carbon · 0.0%

Gap5.5 behindReported by GoogleCounts3% ref. weight
Math2 rows
Math benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
FrontierMath v2 (Tiers 1-3)FrontierMath v2 Tiers 1-368.421%

Best verified: Claude Fable 5.2 · 95.088%

Gap26.7 behindRun by FrontierMath v2 (Tiers 1-3)Counts11% ref. weight
FrontierMath v2 (Tier 4)FrontierMath v2 Tier 421.951%

Best verified: Claude Fable 5.2 · 100.000%

Gap78 behindRun by FrontierMath v2 (Tier 4)Counts2% ref. weight
External signals22 rows
External signals benchmark results, best verified comparison, who ran each test, and whether it counts toward a rank
Game Arena Go1170

Best verified: Claude Fable 5.2 · 1877

Gap707 behindRun by Game Arena GoShown only
ASCII Arena1655

Best verified: Gemini 4 Carbon · 2005

Gap350 behindRun by ASCII ArenaShown only
Landmark Arena1346

Best verified: Claude Opus 5.5 · 1781

Gap435 behindRun by Landmark ArenaShown only
Pelican Arena1563

Best verified: Gemini 4 Neon · 1808

Gap245 behindRun by Pelican ArenaShown only
FrontierPhysicsEnd-to-End Physics Research Benchmark for AI Agents0.0%

Best verified: Claude Fable 5.2 · 33.3%

Gap33.3 behindRun by FrontierPhysicsShown only
NYT ConnectionsNYT Connections LLM Benchmark (extended version)97.4%

Best verified: Gemini 4 Carbon · 99.3%

Gap1.9 behindRun by NYT ConnectionsShown only
Joke Arena1539

Best verified: Gemini 3.1 Pro · 1567

Gap28 behindRun by Joke ArenaShown only
Chart Arena1473

Best verified: GPT-6 Astra · 1614

Gap141 behindRun by Chart ArenaShown only
Landing Page Arena1360

Best verified: GPT-6.1 Sol · 1716

Gap356 behindRun by Landing Page ArenaShown only
Code2Video1440

Best verified: Claude Opus 5.5 · 1585

Gap145 behindRun by Code2VideoShown only
FailureSensorIQ69.0%

Best verified: GPT-6 Astra · 69.6%

Gap0.6 behindRun by FailureSensorIQShown only
DAYJOB Healthcare1.2%

Best verified: Claude Opus 5.5 · 25.2%

Gap24 behindRun by DAYJOB HealthcareShown only
DAYJOB2.6%

Best verified: Claude Opus 5.5 · 24.5%

Gap21.9 behindRun by DAYJOBShown only
FACTS Multimodal57.6%

Best verified: Gemini 4 Carbon · 60.2%

Gap2.6 behindRun by FACTS MultimodalShown only
StealthMarkStealthMark Intelligence Benchmarks (overall score)47.9

Best verified: GPT-6 Astra · 74.6

Gap26.7 behindRun by StealthMarkShown only
BabaIsBenchBabaIsBench (Baba Is You intro levels, pass@1)100%

Best verified: Claude Opus 5 · 100%

GapBest verifiedRun by BabaIsBenchShown only
MaxxBenchMaxxBench (mutated DeepSWE tasks)53.4%

Best verified: Claude Opus 5.5 · 76.4%

Gap23 behindRun by MaxxBenchShown only
Chess PuzzlesChess Puzzles (Epoch AI)58.0%

Best verified: Gemini 4 Neon · 83.0%

Gap25 behindRun by Chess PuzzlesShown only
Mystery Game PuzzlesMystery Game Puzzles (Epoch AI)47.0%

Best verified: GPT-6.2 Astra · 95.0%

Gap48 behindRun by Mystery Game PuzzlesShown only
Furniture AssemblyFurniture Assembly Benchmark (Epoch AI)31.7%

Best verified: Claude Fable 5.2 · 90.0%

Gap58.3 behindRun by Furniture AssemblyShown only
MirrorCodeMirrorCode (Epoch AI and METR)38.9%

Best verified: Claude Sonnet 5.5 · 97.8%

Gap58.9 behindRun by MirrorCodeShown only
HoneyBenchHoneyBench v0.2 (Goodhart Labs)80.5%

Best verified: Claude Fable 5.2 · 11.9%

Gap68.6 behindRun by HoneyBenchShown only

Outlined bars are lab-reported. Filled bars are benchmark-run or independent.

Scores by reasoning effort

Gemini 3.8 Flash has published results at 3 reasoning effort settings on 4 benchmarks, from ARC Prize. Each table is one benchmark run at every effort the publisher tested, with the cost it reported. BenchLM's score uses at most one setting per benchmark; these tables show the rest.

ARC-AGI-1 · Semi-Private set

From Low to High effort, the score goes from 90.5% to 98.5% and the cost per task from $0.067 to $0.21.

ARC-AGI-1, Semi-Private set: score and cost per task at each reasoning effort
EffortScoreCost per task
Low90.5%$0.067
Medium97.5%$0.11
Hightop score98.5%$0.21

Source: ARC Prize.

ARC-AGI-2 · Semi-Private set

From Low to High effort, the score goes from 77.5% to 89.2% and the cost per task from $0.15 to $0.40.

ARC-AGI-2, Semi-Private set: score and cost per task at each reasoning effort
EffortScoreCost per task
Low77.5%$0.15
Medium82.9%$0.25
Hightop score89.2%$0.40

Source: ARC Prize.

ARC-AGI-3 · Standard harness

From Low to High effort, the score goes from 6.0% to 10.4%.

ARC-AGI-3, Standard harness: score at each reasoning effort
EffortScore
Low6.0%
Medium4.0%
Hightop score10.4%

Source: ARC Prize.

ARC-AGI-3 (Provider Adapter) · Provider Adapter harness

From Low to High effort, the score goes from 15.1% to 35.0%.

ARC-AGI-3 (Provider Adapter), Provider Adapter harness: score at each reasoning effort
EffortScore
Low15.1%
Medium24.2%
Hightop score35.0%

Source: ARC Prize.

Publisher scores and costs · display only, not part of the BenchLM score

How the score is built

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.

Current modelExplore all models

Gemini 3.8 Flash · 67.9 score · $2.25 blended per million tokens

The chart opens on the current model. Scroll horizontally to inspect the full price axis.

30405060708090100$0.10$0.50$1$5$10$25↘ frontierGemini 3.8 Flash

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.

0352026-092026-10Output priceInput price

2 dated first-party price records. The newest record lists $3.75 per million tokens.

Lineage

The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.

  1. May 19, 2026

    Gemini 3.5 Flash

    Score 59.9 · $1.5 / $9

  2. Jul 21, 2026

    Gemini 3.6 Flash

    Score 61.6 · $0.75 / $3.75

  3. Aug 13, 2026

    Gemini 3.7 Flash

    Score 66.6 · $0.75 / $3.75

  4. Sep 2, 2026 · you are here

    Gemini 3.8 Flash

    Score 67.9 · $0.75 / $3.75

  5. Nov 12, 2026

    Gemini 4 Hydrogen

    Score 78.8 · $1.5 / $7.5

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-3.8-flashGoogle Gemini 3.8 Flash API documentation
Maximum output
Not sourced yet
Knowledge cutoff
Not sourced yet
Input modalities
text, image, video, audio, pdfGoogle Gemini 3.8 Flash API documentation
Parameters
Not disclosed by the provider
Availability
Gemini API · Google AI Studio · Android Studio · Stitch · Gemini app · Gemini Enterprise · Google AI Mode · Google Sheets · Google AntigravityGoogle Gemini 3.8 Flash and Gemini 3.8 Flash Cyber launch
Cloud regions
Not tracked yet
API capabilities
Tool calling, structured outputs, and batch support are not tracked yet
Prompt caching
Published at $0.075 per million cached input tokensGoogle Gemini API pricing
Self-host
Weights are not published
Rate limits
Not tracked yet

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 3.8 Flash ranks #28 of 216 on the public leaderboard with a score of 67.9/100. Its source-verified position is #27 of 147.

Gemini 3.8 Flash 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.

Generally available as the stable `gemini-3.8-flash` model through the Gemini API and Google AI Studio. Google also distributes it through Android Studio, Stitch, the Gemini app, Gemini Enterprise, Google AI Mode, Google Sheets, and Google Antigravity.

Gemini 3.8 Flash has cybersecurity benchmark results. The cybersecurity benchmark comparison shows them beside other models, one security job at a time, with the access route for restricted models.

Gemini 3.8 Flash sits in the Gemini 3.8 Flash family with Gemini 3.8 Flash Cyber. Its explicit predecessor is Gemini 3.7 Flash. 59 of 952 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.

Its strongest eligible category is Multilingual at #13, while its lowest eligible position is Coding at #36. a well-rounded choice across a range of tasks.

Last updated November 21, 2026. Runtime fields remain blank until a sourced snapshot exists.

Deployment options

Self-host and provider-specific paths stay separate from benchmark evidence so operating constraints are visible before a score becomes the whole decision.

Where to run Gemini 3.8 Flash by API

OpenRouter is the one host we track that publishes a per-token price for it. Prices read from each host’s own pages on November 16, 2026.

Inference hosts for Gemini 3.8 Flash with input, output and cached-input prices per million tokens.
HostInput /1MOutput /1MCached /1MContext
OpenRouter$0.82$4.13$0.0831.05M
Compare inference hosts

Questions

How does Gemini 3.8 Flash perform overall in AI benchmarks?

Gemini 3.8 Flash ranks #28 out of 216 models on the public BenchAlign leaderboard, with a score of 67.9/100. Its evidence status is Supported, and this profile shows 59 source-displayable benchmark rows. The label describes evidence depth, not a provider quality claim; inspect category rows before choosing a workload.

Is Gemini 3.8 Flash good for knowledge and understanding?

Gemini 3.8 Flash ranks #16 out of 158 eligible models for knowledge and understanding, with a public category score of 71.5/100. 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 3.8 Flash good for coding and programming?

Gemini 3.8 Flash ranks #36 out of 180 eligible models for coding and programming, with a public category score of 65.5/100. 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 3.8 Flash good for mathematics?

Gemini 3.8 Flash ranks #24 out of 125 eligible models for mathematics, with a public category score of 64.3/100. 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 3.8 Flash good for reasoning and logic?

Gemini 3.8 Flash ranks #24 out of 161 eligible models for reasoning and logic, with a public category score of 69.3/100. 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 3.8 Flash good for agentic tool use and computer tasks?

Gemini 3.8 Flash ranks #31 out of 182 eligible models for agentic tool use and computer tasks, 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 3.8 Flash good for multimodal and grounded tasks?

Gemini 3.8 Flash ranks #23 out of 119 eligible models for multimodal and grounded tasks, with a public category score of 66.8/100. 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 3.8 Flash good for instruction following?

Gemini 3.8 Flash ranks #21 out of 143 eligible models for instruction following, with a public category score of 65.6/100. 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.

Which sibling models are related to Gemini 3.8 Flash?

Gemini 3.8 Flash belongs to the Gemini 3.8 Flash family. Related tracked variants include Gemini 3.8 Flash Cyber. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.

Does Gemini 3.8 Flash have full benchmark coverage on BenchLM?

No. Gemini 3.8 Flash currently has 82 source-displayable rows. 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 3.8 Flash?

Gemini 3.8 Flash 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 3.8 Flash with every tracked model1034 comparisons