Google · in Borg

Gemini 3.8 Flash

Gemini 3.8 Flash is Google's newer Flash workhorse, building on 3.7 Flash with a focus on long-horizon software engineering and autonomous, multi-step knowledge workflows.

Reasoning1,048,576 tokenstext, image, video, audio, pdf

1,048,576 tokensContext window

2510 msMedian time to first token (gateway proxy)

29.2Output tokens per second (whole turn)

$0.75 / $3.75Vendor API price in / out · per 1M

Included in Borg

Download Borg Borg for Business

At a glance

Specification	Value
Provider	Google
Family	Gemini
Release date	2026-09-02 Vendor source
Context window	1,048,576 tokens Vendor source
Input modalities	text, image, video, audio, pdf
Output modalities	text
Reasoning	Yes
Open weights	No
Borg lane ids	`antigravity/gemini-3.8-flash-high`, `antigravity/gemini-3.8-flash-low`, `antigravity/gemini-3.8-flash-medium`, `antigravity/gemini-3.8-flash-tiered`
Versions in Borg	Gemini 2.5 Flash · Gemini 3 Flash · Gemini 3.5 Flash · Gemini 3.6 Flash · Gemini 3.7 Flash · Gemini 3.8 Flash
Median time to first token (gateway proxy) on Borg	2510 ms
Median tokens per second (end to end) on Borg	29.2 output tokens per second

Not published by the vendor: thinking modes, knowledge cutoff, fusion roles.

What Gemini 3.8 Flash is good for

Gemini 3.8 Flash is Google's newer Flash workhorse, building on 3.7 Flash with a focus on long-horizon software engineering and autonomous, multi-step knowledge workflows. It retains the speed-and-cost positioning of Flash rather than the heavier Pro tier, while supporting low, medium and high thinking effort instead of minimal. Its multimodal input window can hold text, code, images, video, audio and PDFs, though responses are text, not newly generated images. For basic classification the cheaper Lite models are a better match; this one is intended for harder loops that need planning and follow-through. Borg includes 3.8 Flash in the subscription; Borg Fusion can route demanding agent steps to it, or you can pick an effort-specific lane directly.

long-horizon coding agents
enterprise document workflows
autonomous multi-step tool use

How Borg uses it

Gemini 3.8 Flash is available as a subscription catalog lane in Borg. The catalog does not publish which Fusion profiles route to this model, so no routing role is claimed here. You can pick it directly by setting the OpenAI-compatible gateway model field to a listed lane id. Check the current catalog for availability before building a workflow around a lane.

model: "antigravity/gemini-3.8-flash-high"

Measured on Borg

Metric	Median	95th percentile	Samples
Time to first token (gateway proxy)	2510 ms	7172 ms	1147
Tokens per second (end to end)	29.2 output tokens per second	169.6 output tokens per second	85551

Time to first token (gateway proxy)

Lower is fasterFleet median 3971.5 ms

Gemini 3.8 Flash2510 ms

Output tokens per second (whole turn)

Higher is fasterFleet median 31.9 tokens/s

Gemini 3.8 Flash29.2 tokens/s

Time to first token (gateway proxy): time until the first streamed response event reaches the gateway; our closest available measure of time to first token
Tokens per second (end to end): tokens per second, end to end: provider-reported output tokens divided by the whole turn's elapsed time, including queueing and reasoning; not a streaming per-token rate

Completion speed covers the whole turn including queueing and reasoning time; it is not a streaming tokens-per-second figure.

Measured on Borg: SearchAtlas AI Gateway production request logs. 30-day window ending 2026-10-08; 92815 requests. Minimum 100 samples per published request metric.

Benchmarks

Google-reported internal automated safety evaluations, relative to the comparator named below. These are changes in evaluation performance, not absolute capability scores or human red-team results. Google says updated evaluations are not directly comparable with results in earlier model cards. The vendor table reports percentage-point change: lower is better for safety and unjustified-refusal measures; higher is better for tone. Vendor evaluation details

Benchmark	Vendor-reported score	Source
Text-to-text safety vs Gemini 3.7 Flash	-0.4pp	Vendor source
Multilingual safety vs Gemini 3.7 Flash	+5.4pp	Vendor source
Tone vs Gemini 3.7 Flash	+0.2pp	Vendor source

Pricing

Included in Borg subscription. Vendor API list price: $0.75 input / $3.75 output per 1M tokens. Vendor price source.

Introductory rates through December 31, 2026. From January 1, 2027: $1.50 input / $7.50 output / $0.15 cached input per 1M tokens. Vendor pricing terms

Compare subscription and API pricing · Cost Savings Calculator

FAQ

What is Gemini 3.8 Flash?

Gemini 3.8 Flash is listed in Borg's subscription catalog as antigravity/gemini-3.8-flash-high. Gemini 3.8 Flash is Google's newer Flash workhorse, building on 3.7 Flash with a focus on long-horizon software engineering and autonomous, multi-step knowledge workflows.

Is Gemini 3.8 Flash in Borg?

Yes. Gemini 3.8 Flash is listed in Borg's subscription catalog under the lane antigravity/gemini-3.8-flash-high. You can choose the lane directly, or explore the available Fusion profiles. Availability and model details can change; check the current catalog before deployment.

What context window does Gemini 3.8 Flash have?

The published context window is 1,048,576 tokens. Consult the cited sources for details and current limits. Vendor source

Is Gemini 3.8 Flash good for coding?

Vendor-published evaluations appear in the benchmarks section; their settings may differ from your tasks. Its listed uses include long-horizon coding agents, enterprise document workflows, autonomous multi-step tool use. Try the lane for your own tasks, and compare the output with your requirements before choosing a default.

How fast is Gemini 3.8 Flash on Borg?