Effort: high · High
GoogleReleased Sep 2, 2026
Gemini 3.8 Flash
Measured performance
All leaderboards ↗Benchmark results
Effort: high · High
GDPval-AA ↗2.1
1,436.3Effort: high · High
AutomationBench-AA ↗AA Index 4.3.2 collection
59.93%Effort: high · High
Effort: high · High · Harness unreported
SciCode ↗AA Index 4.3.2 collection
56.6%AA · Under reviewEffort: high · High
Humanity’s Last Exam ↗AA Index 4.3.2 collection
47.82%Effort: high · High
GDP.pdf ↗AA Index 4.3.2 collection
21%Effort: high · High
CritPt ↗AA Index 4.3.2 collection
18.29%AA · Under reviewEffort: high · High
AA-Omniscience ↗AA Index 4.3.2 collection
29.55Effort: high · High
AA-LCR ↗1.1
81.33%Effort: high · High
Provider measurements
1 configurationSpeed
≈10K input tokens1 requestmedian / 72h
Artificial Analysis · first-party providers · serving endpoint not reported
Gemini 3.8 Flash (High)
Reasoning high
- Output rate
- 148.729 tok/s
- Time to first chunk
- 26.2474 s
Last checked Source ↗
Methodology 2.2.0 & measurement details
First chunk includes request processing and network time, and may include reasoning.
- Prompt
- long · ≈10,000 input tokens
- Parallel requests
- 1
- Minimum answer
- 1,500 tokens
- Aggregation
- median over 72 hours
- Token counting
- o200k_base
- Benchmark client
- Google Cloud us-central1-a
- Serving endpoint
- Not reported
- Serving region
- Not reported
- Sample size
- Not reported
- Window end
- Not reported
Access & cost
0 quotesPrices across providers
No listed provider price
Model specifications
- Developer
- Modality
- Language
- Context
- 1,000,000 tokens
- Maximum output
- Not reported
- Weights
- Closed weights
- Released
- Sep 2, 2026
- First discovered
- Oct 16, 2026
Exact identities
- artificial-analysis
8a999846-4c1d-4ce7-a8b7-1310a7166fd7
Sources
Updated Oct 20, 2026
More from Google
View all ↗Price not listed
Price not listed
Price not listed