Gemini 3.8 Flash or Gemini 3 Flash Preview: which to choose?
Gemini 3.8 Flash (Google DeepMind) and Gemini 3 Flash Preview (Google DeepMind) compared point by point: measured performance, pricing, context and capabilities.
Data as of October 22, 2026
Side-by-side comparison
| Criterion | Gemini 3.8 Flash | Gemini 3 Flash Preview |
|---|---|---|
| Developer | Google DeepMind | Google DeepMind |
| Released | September 2, 2026 | December 17, 2025 |
| Capability index | 83.7 / 100 | 80.3 / 100 |
| Input price | $0.75 | $0.50 |
| Output price | $3.75 | $3 |
| Cost of a typical task | $0.023 | $0.016 |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Maximum output | 65,536 tokens | 65,536 tokens |
| Reasoning mode | Yes | Yes |
| Tool calling | Yes | Yes |
| Accepted inputs | text, image, video, audio, PDF | text, image, video, audio, PDF |
| Open weights | No | No |
| Knowledge cutoff | Not disclosed | January 2025 |
The match-up at a glance
Gemini 3.8 Flash and Gemini 3 Flash Preview are two mid-range models, balancing quality and cost, both built by Google DeepMind. Gemini 3.8 Flash was released on September 2, 2026, Gemini 3 Flash Preview on December 17, 2025.
All figures below come from public sources checked on October 22, 2026: models.dev and OpenRouter for prices and capabilities, Epoch AI for scores.
Performance
On our capability index, Gemini 3.8 Flash scores 83.7 out of 100 and Gemini 3 Flash Preview 80.3. Gemini 3.8 Flash leads by 3.4 points.
On the 3 benchmarks reported for both, Gemini 3.8 Flash wins 3 times and Gemini 3 Flash Preview 0:
- GPQA Diamond (PhD-level science questions): Gemini 3.8 Flash 95.4%, Gemini 3 Flash Preview 89.4%
- OTIS Mock AIME (competition mathematics problems): Gemini 3.8 Flash 98.9%, Gemini 3 Flash Preview 95.6%
- FrontierMath (very hard research-level mathematics): Gemini 3.8 Flash 68.4%, Gemini 3 Flash Preview 51.2%
Pricing
Gemini 3.8 Flash costs $0.75 per million input tokens and $3.75 per million output tokens; Gemini 3 Flash Preview costs $0.50 and $3.
For a typical agent task (read 20,000 tokens, write 2,000), expect $0.023 with Gemini 3.8 Flash versus $0.016 with Gemini 3 Flash Preview. Gemini 3 Flash Preview is 1.4 times cheaper, a gap that adds up quickly over 1,000 tasks: $6.5 difference.
Run the numbers with your own volumes in the agent cost calculator.
Under the hood
Both models have the same context window: 1,048,576 tokens, roughly 1,573 pages.
Reasoning mode: Gemini 3.8 Flash yes, Gemini 3 Flash Preview yes. Tool calling: Gemini 3.8 Flash yes, Gemini 3 Flash Preview yes. Image input: Gemini 3.8 Flash yes, Gemini 3 Flash Preview yes.
Neither publishes its weights: both are used through an API.
In short
Our reading of the data:
- Choose Gemini 3.8 Flash for a higher capability index and a more recent release.
- Choose Gemini 3 Flash Preview for a lower cost.
This verdict relies on public figures; it does not replace a trial on your own tasks.
Other comparisons
Frequently asked questions
Gemini 3.8 Flash or Gemini 3 Flash Preview: which is better?
On our index (Epoch AI data), Gemini 3.8 Flash scores 83.7 and Gemini 3 Flash Preview 80.3 out of 100. The best choice also depends on price and on your use case.
Which one is cheaper?
Gemini 3 Flash Preview. To read 20,000 tokens and write 2,000, expect $0.023 with Gemini 3.8 Flash and $0.016 with Gemini 3 Flash Preview (prices as of October 22, 2026).
Which one has the larger context window?
Neither: both accept 1,048,576 tokens.
Can I run Gemini 3.8 Flash or Gemini 3 Flash Preview on my own servers?
No: both are closed models, available through an API.
Data sources
- Models.dev — Model data: models.dev (open-source project)
- OpenRouter — Market prices cross-checked with OpenRouter
- Epoch AI — AI Benchmarking Hub — Epoch AI, “AI Benchmarking Hub”, epoch.ai, CC BY 4.0
Data as of October 22, 2026
Report an error
Does a figure, a date or a fact look wrong? Tell us: every report is checked.