Gemini 3.8 Live

Google · Released Sep 2026

Conditional

Gemini 3.8 Live is Google's default Live API model for low-latency voice agents that keep talking while tool calls finish in the background. Conditional because Google bills Gemini 3.8 Live Extended Thinking at the identical rate, and Artificial Analysis scores that sibling higher on its Speech to Speech Index, so you pay the same for less reasoning. Pick 3.8 Live only when turn latency matters more than reasoning depth: developers shipping real-time voice experiences on the Gemini API.

Is it right for you?

Good for

Voice agents where turn latency is the priority, per Google's "without reasoning-induced delays" positioning
Keeping a conversation going while tools and API calls finish in the background

Not good for

Multi-step reasoning mid-call, where Google recommends the Extended Thinking model
Workloads that need structured outputs, context caching, code execution or the Batch API, none of which the model supports
Sensitive data on the free tier, which Google's pricing page marks as used to improve its products

How it performs by task

Low-latency voice conversation

Very Good

Google's default Live model for low-latency voice without reasoning-induced delays

Complex reasoning during calls

Fair

Google recommends the Extended Thinking model when higher background reasoning is required

Pricing

Input

$0.75 / 1M (text)

Output

$4.50 / 1M (text)

Context

131K in, 65K out; audio $3/$12 per 1M

View full pricing

Benchmarks

Benchmark	Score	Source
Artificial Analysis Speech to Speech Index	76.0	Source

No verdict changes yet

The clock starts day one. Changes land here as our verdict evolves.

Sources

Verification log

Pricing· No changesOct 5, 2026
Automated agent
Pricing· No changesSep 30, 2026
Automated agent

How we evaluate