Google
·Released on Sep 2, 2026

Gemini 3.8 Flash API Benchmarks, Pricing & Provider Data

Compare Gemini 3.8 Flash with another model

Choose a model to open its comparison page.

Share on X
LLM

Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.

Category Performance

Observed-capability estimates with uncertainty reported separately from independent benchmark families.

Coverage
2 / 8
Methodology
V3.0
Category PerformanceObserved-capability estimates with uncertainty reported separately from independent benchmark families.Agents-Coding61.5Reasoning62.4Knowledge-Math-Multilingual-Multimodal-Instruction following-
#1Reasoning62.4Provisional1/4 Measured dimensions
80% interval: 48.576.3
abstract logicPrior only
scientific causal62.4
gpqa · gpqa / hle · hle
multistep constraintsPrior only
evidence verificationPrior only
#2Coding61.5Provisional1/4 Measured dimensions
80% interval: 45.577.4
code generation61.5
scicode · scicode
repository engineeringPrior only
debugging testingPrior only
tooling qualityPrior only
No data:AgentsKnowledgeMathMultilingualMultimodalInstruction following

Specifications

Input and output token limits for this model, plus how it ranks on long-context understanding.

INPUT
1.0Mtokens
1.3K pages of text
OUTPUT
65.5Ktokens
8K128K1M4M
1.0M

Features

Technical Details

Input
Output
Released
Sep 2026
Documentation
Tokenizer
Gemini
Architecture
text+image+file+audio+video->text
Moderated
No
Supported parameters
include_reasoningmax_tokensreasoningreasoning_effortresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_p

Rankings

Excels at

Detailed scores

Updated: Sep 3, 2026

Speed & latency

undefined metric} other undefined metrics}}

Output speed297.5 tok/s#5 / 79Time to first token10.03 s#69 / 79
Output speed
LMSpeed rank#5
Score297.5 tok/s
UpdatedSep 3, 2026
Confidence4
Time to first token
LMSpeed rank#69
Score10.03 s
UpdatedSep 3, 2026
Confidence4

Pricing

undefined metric} other undefined metrics}}

Input price$0.750/M#105 / 183Output price$3.75/M#108 / 183
Input price
LMSpeed rank#105
Score$0.750/M
UpdatedSep 3, 2026
Confidence4
Output price
LMSpeed rank#108
Score$3.75/M
UpdatedSep 3, 2026
Confidence4

Agents

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Planning & decompositionPrior only
Tool usePrior only
Environment & long-horizon executionPrior only
Recovery & completion reliabilityPrior only

Coding

V3.0

undefined metric} other undefined metrics}} · Provisional

Score61.580% interval45.5–77.41/4 Measured dimensionsSciCode53.6%#15 / 208
Dimensions and evidence
Code generation61.5
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
scicode · scicode · z 1.40 · q 1.00
Repository engineeringPrior only
Debugging & testingPrior only
Tool-assisted development & qualityPrior only
SciCodeV3 evidence
LMSpeed rank#15
Score53.6%
UpdatedSep 3, 2026
Confidence4

Reasoning

V3.0

undefined metric} other undefined metrics}} · Provisional

Score62.480% interval48.5–76.31/4 Measured dimensionsGPQA95.3%#1 / 215HLE47.8%#9 / 212
Dimensions and evidence
Abstract logicPrior only
Scientific & causal reasoning62.4
undefined family} other undefined families}} · undefined metric} other undefined metrics}}
gpqa · gpqa · z 1.36 · q 1.00
hle · hle · z 0.99 · q 1.00
Multi-step constraintsPrior only
Evidence integration & verificationPrior only
GPQAV3 evidence
LMSpeed rank#1
Score95.3%
UpdatedSep 3, 2026
Confidence4
HLEV3 evidence
LMSpeed rank#9
Score47.8%
UpdatedSep 3, 2026
Confidence4

Knowledge

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Broad knowledgePrior only
Professional knowledgePrior only
FactualityPrior only
Retrieval & open-book usePrior only

Math

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Foundational mathPrior only
Competition mathPrior only
Advanced proofsPrior only
Applied & tool-assisted mathPrior only

Multilingual

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Cross-language understandingPrior only
Multilingual generationPrior only
Reasoning transferPrior only
Low-resource robustnessPrior only

Multimodal

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Perception & OCRPrior only
Document & spatial understandingPrior only
Visual reasoningPrior only
Video & grounded actionPrior only

Instruction following

V3.0

undefined metric} other undefined metrics}} · No data

No data80% interval30.8–69.20/4 Measured dimensions
Dimensions and evidence
Constraint followingPrior only
Structured outputPrior only
Novel-instruction generalizationPrior only
Multi-turn & long instructionsPrior only

Rankings are based on community-submitted tests and periodic health probes. Advisory only, not official data.Artificial AnalysisStandard benchmark data may include BenchLM and other public sources.

Build with LMSpeed Data

Free

Free public API for LLM pricing, benchmarks & provider data

  • Real-time pricing & availability
  • Speed & latency benchmarks
  • 300+ models, 600+ providers
  • 1,000 requests/day
View API Documentation