Data as of October 29, 2026 · How the score is built
Gemini 3.8 Live
Decision readingGemini 3.8 Live is tracked, but not publicly ranked yet. Published speech or voice results from Gemini 3.8 Audio model evaluation appear below. They do not produce a weighted text-model score.
Released Sep 15, 2026 — see all recent releases
Voice benchmark evidence
Gemini 3.8 Audio model evaluation is an owner-defined voice evaluation. Its result stays separate from BenchLM's weighted text-model ranking.
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Base entry
Gemini 3.8 Live Extended ThinkingSpec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- gemini-3.8-liveGoogle Gemini 3.8 Live model documentation
- Context window
- 128KGoogle DeepMind Gemini 3.8 Audio model card
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- audio, text, image, videoGoogle DeepMind Gemini 3.8 Audio model card
- Output modalities
- audio, textGoogle DeepMind Gemini 3.8 Audio model card
- Parameters
- Not disclosed by the provider
- Availability
- Gemini API · Google AI Studio · Vertex AIGoogle DeepMind Gemini 3.8 Audio model card
- Cloud regions
- Not tracked yet
- Lifecycle
- activeGoogle Gemini 3.8 Live model documentation
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Not documented in the pricing recordGoogle model documentation
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track Gemini 3.8 Live, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.
Gemini 3.8 Live is a proprietary model with a 128K context window. No explicit reasoning mode is documented in this profile.
Available as the stable `gemini-3.8-live` Live API model, with Google AI Studio, Gemini app, Vertex AI, and Google Search Live access listed on the model card. The launch announcement adds real-time troubleshooting inside Search Live, a private preview in Gemini Enterprise, automatic detection and mid-conversation switching across 97 supported languages, and near real-time visual input. All audio output is watermarked with SynthID.
Google’s September 15 Gemini 3.8 Audio card describes the default Live model with 128K input context and audio and text output. Its voice evaluations stay separate from weighted text-model scores; Extended Thinking results are not attributed to this row.
Gemini 3.8 Live sits in the Gemini 3.8 Live family with Gemini 3.8 Live Extended Thinking. Its explicit predecessor is Gemini 3.1 Flash Live Preview. Speech or voice evidence appears above; text-model benchmark coverage remains empty.
Last updated October 29, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does Gemini 3.8 Live perform overall in AI benchmarks?
Gemini 3.8 Live has published speech or voice results in Gemini 3.8 Audio model evaluation. Those results remain separate from the weighted text-model ranking, so this profile does not assign a BenchLM score or overall rank. Use the linked evaluation's metric and run conditions to compare speech systems.
Which sibling models are related to Gemini 3.8 Live?
Gemini 3.8 Live belongs to the Gemini 3.8 Live family. Related tracked variants include Gemini 3.8 Live Extended Thinking. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does Gemini 3.8 Live have full benchmark coverage on BenchLM?
No. Gemini 3.8 Live's speech or voice results appear in Gemini 3.8 Audio model evaluation, but it has no published text-model benchmark rows. The voice evaluation does not fill the text-model coverage slots or produce an overall rank. Missing text coverage is not treated as a zero score.
What is the context window size of Gemini 3.8 Live?
Gemini 3.8 Live has a documented context window of 128K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.