Google model product
Gemini 3.8 Flash
8 Flash is a Google LLM that accepts text, image, video, audio, and PDF input and produces text.
Updated Sep 8, 2026. Default version: Gemini 3.8 Flash
LLMBoard Score86.7Gemini 3.8 Flash
Coverage0%9 benchmark families
Context window1MTokens
Official input price$0.75Google API
Gemini 3.8 Flash Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Gemini 3.8 Flash Benchmark Results
Benchmark scores for Gemini 3.8 Flash.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkHarvey's Legal Agent Benchmark | Score10.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkHLE-Verified | Score54.90% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLABBench2 | Score86.20% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLVBench | Score87.10% | Rank01 | Participants28 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.1 | Score89.40% | Rank01 | Participants35 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVals Finance Agent v2 | Score61.40% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGDP.pdf | Score35.00% | Rank02 | Participants7 | Percentile83.33% | EvidenceC | Evaluated |
| BenchmarkBioMysteryBench | Score56.50% | Rank03 | Participants5 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkDeepSWE 1.1 | Score73.70% | Rank03 | Participants31 | Percentile93.33% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,489.89 rating | Rank03 | Participants121 | Percentile98.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,495.23 rating | Rank04 | Participants210 | Percentile98.56% | EvidenceA | Evaluated |
| BenchmarkOSWorld 2.0 | Score59.00% | Rank06 | Participants11 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score14.25% | Rank06 | Participants49 | Percentile89.58% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,493.85 rating | Rank06 | Participants210 | Percentile97.61% | EvidenceA | Evaluated |
| BenchmarkGDPval-AA | Score1,545.00 points | Rank07 | Participants9 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score10.42% | Rank09 | Participants49 | Percentile83.33% | EvidenceA | Evaluated |
| BenchmarkTerminal-Bench 4.0 | Score19.10% | Rank10 | Participants13 | Percentile25.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score5.47% | Rank14 | Participants49 | Percentile72.92% | EvidenceA | Evaluated |
| BenchmarkCharXiv-R | Score86.20% | Rank16 | Participants55 | Percentile72.22% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,567.23 rating | Rank19 | Participants97 | Percentile81.25% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score3.72% | Rank24 | Participants49 | Percentile52.08% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.70% | Rank25 | Participants49 | Percentile50.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score-1.73% | Rank32 | Participants49 | Percentile35.42% | EvidenceA | Evaluated |
Gemini 3.8 Flash Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categorymulti turn | Rank01 | Rating / score1,512.22 | Votes779 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycreative writing | Rank02 | Rating / score1,481.06 | Votes1,034 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank03 | Rating / score1,558.77 | Votes354 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank03 | Rating / score1,497.28 | Votes1,086 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank03 | Rating / score1,523.73 | Votes256 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryenglish | Rank03 | Rating / score1,497.90 | Votes2,002 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexclude ties | Rank03 | Rating / score1,505.05 | Votes3,775 | ObservationsN/A | Result date |
| Arenatext factuality | Categorylonger query | Rank03 | Rating / score1,509.45 | Votes2,577 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank03 | Rating / score1,489.89 | Votes5,125 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank03 | Rating / score1,522.40 | Votes256 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank04 | Rating / score1,502.06 | Votes2,002 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank04 | Rating / score1,510.85 | Votes3,775 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank04 | Rating / score1,512.42 | Votes1,355 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank04 | Rating / score1,494.72 | Votes1,032 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry writing and literature and language | Rank04 | Rating / score1,500.93 | Votes1,391 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank04 | Rating / score1,495.23 | Votes5,125 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycoding | Rank04 | Rating / score1,537.55 | Votes1,447 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts english | Rank04 | Rating / score1,518.57 | Votes1,331 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry entertainment and sports and media | Rank04 | Rating / score1,469.04 | Votes1,374 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank04 | Rating / score1,494.90 | Votes1,086 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank04 | Rating / score1,499.45 | Votes1,391 | ObservationsN/A | Result date |
| Arenatext style control | Categorymulti turn | Rank04 | Rating / score1,515.94 | Votes779 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank05 | Rating / score1,508.38 | Votes3,458 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank05 | Rating / score1,506.78 | Votes2,577 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts | Rank05 | Rating / score1,511.60 | Votes3,458 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank05 | Rating / score1,497.54 | Votes975 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry writing and literature and language | Rank05 | Rating / score1,488.32 | Votes1,373 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank05 | Rating / score1,547.07 | Votes354 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank05 | Rating / score1,537.24 | Votes1,456 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank05 | Rating / score1,503.88 | Votes2,002 | ObservationsN/A | Result date |
Gemini 3.8 Flash Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.75 input, $3.75 output per 1M
- Official provider
- Lowest third-party
- From $0.75 input, $3.75 output per 1M via NanoGPT
- Tracked offerings
- 15
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNanoGPT | Provider model IDgoogle/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderVertex | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderGoogle | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDgoogle/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDgoogle/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderRequesty | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDgoogle/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDgoogle/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderEden AI | Provider model IDgoogle/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderCrossModel | Provider model IDgemini/gemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderGitHub Copilot | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderVivgrid | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$0.75 | Output / 1M$3.75 | Context1M | Updated |
| ProviderVenice AI | Provider model IDgemini-3-8-flash | Regionglobal | Input / 1M$0.9375 | Output / 1M$4.69 | Context1M | Updated |
| ProviderOpenCode Zen | Provider model IDgemini-3.8-flash | Regionglobal | Input / 1M$1.5 | Output / 1M$7.5 | Context1M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Gemini 3.8 Flash Runtime Performance
Provider-specific output speed and catalog latency for Gemini 3.8 Flash. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderGoogle | Output Speed51.56 tok/s | Catalog Latency11.37 s | Max Input1M | Max Output65.5K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Gemini 3.8 Flash Specifications
Technical details for the model's default version.
- Version
- Gemini 3.8 Flash
- Released
- Sep 2, 2026
- Knowledge cutoff
- Mar 31, 2026
- Parameters
- N/A
- Context window
- 1M
- Max output
- 65.5K
- Inputs
- audio, image, text, video
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Gemini 3.8 Flash Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemini 3.8 Flash | Released | LLMBoard86.74 | ParametersN/A | Context1M | Max output65.5K | Open weightsNo | LicenseProprietary |
Models similar to Gemini 3.8 Flash
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
GO
Google
Gemini 3.7 Flash
GOGoogle
Gemini 3.5 Flash
GOGoogle
Gemini 3.1 Pro
GOGoogle
Gemini 3.6 Flash
GOGoogle
Gemini 3.8 Flash Cyber
GOGoogle
Gemini 3 Pro
What is Gemini 3.8 Flash?
Key information about Gemini 3.8 Flash and its available data.
8 Flash is a Google large language model for software engineering, autonomous agents, and enterprise workflows. It supports low, medium, and high thinking levels, with a 1,048,576-token input context and 65,536-token output.
Data as of 2026-09-08.
FAQ
Common questions about Gemini 3.8 Flash.
When was Gemini 3.8 Flash released?
Gemini 3.8 Flash's default version was released on Sep 2, 2026.
How much does Gemini 3.8 Flash cost?
Gemini 3.8 Flash's official API price is $0.75 per million input tokens and $3.75 per million output tokens via Google. The lowest tracked third-party offer starts at $0.75 input and $3.75 output via NanoGPT.
Who created Gemini 3.8 Flash?
Gemini 3.8 Flash was created by Google.
What is the context window for Gemini 3.8 Flash?
The default version has a 1M token context window.
Is Gemini 3.8 Flash open weight?
No. The default version is not marked as having publicly available weights.