Twelve Pages Gemini 3.8 Live & Omni 1.2 Flash Studio

Gemini 3.8 Live & Omni 1.2 Flash Studio

Build low-latency bidirectional voice/vision agents with Gemini 3.8 Live (Extended Thinking, NON_BLOCKING tools, Live Translate) and multi-shot 4K generative video workflows with Gemini Omni 1.2 Flash.

Updated 2026-11 v1.0 12 min read 12-Page Illustrated Edition AI-Authored & Human-Directed
Gemini 3.8 Live soundwaves and Omni 1.2 Flash cinema film frames converging in an obsidian crystal prism

Core Takeaways

Zero-Lag Duplex Voice & Vision

Stream 16kHz PCM audio and JPEG video frames bidirectionally while Extended Thinking speaks natural fillers during async tool calls.

Stateful 4K Cinema Studio

Generate 3s–15s clips, interpolate keyframes, strip and regenerate scores, or extend scenes to 30 seconds over the Interactions API.

Production-Hardened SDKs

Build securely on google-genai >= 2.30.1 and @google/genai >= 2.29.1 with resumption tokens and WebRTC partners.

1. Two Engines, One Unified SDK Mental Model

Modern multimodal apps split cleanly into two complementary rhythms: an always-open WebSocket conduit for sub-second human conversation, and a stateful Interactions loop for asynchronous 4K video creation and multi-turn refinement.

Quick Navigation

A Real-Time Conduit (Live API)

Think of gemini-3.8-live as a live phone call that never hangs up between sentences. You open a persistent WebSocket via client.aio.live.connect, stream raw 16kHz PCM audio and JPEG frames with send_realtime_input, and receive 24kHz PCM voice chunks back in milliseconds—ready for natural barge-in interruptions at any instant.

B Stateful Cinema Studio (Interactions API)

Think of gemini-omni-1.2-flash as a non-linear film suite driven by client.interactions.create. Each request turns prompts, uploaded Files API assets, or a previous_interaction_id into crisp 360p–4K MP4 output—retaining server-side creative history automatically when store=True.

Dual-Engine Architecture Diagram

UNIFIED GOOGLE GENAI SDK (PYTHON >= 2.30.1 / TS >= 2.29.1) STREAMING ENGINE • PERSISTENT WEBSOCKET Gemini 3.8 Live & Extended Thinking • Input: 16kHz PCM audio, JPEG video, text (send_realtime_input) • Output: 24kHz PCM audio + transcripts + async tool calls • Specialists: 3.5 Transcribe Live & 3.5 Live Translate (70+ langs) • Guardrails: interaction_status, GoAway, Ephemeral Tokens CREATIVE ENGINE • STATEFUL INTERACTIONS API Gemini Omni 1.2 Flash Studio • Files API: Poll media until ACTIVE, bind via natural language • Generate: 3s–15s (360p to 4K), keyframes (3s–10s), R2V • Refine: Surgical edits, --strip-audio, extend up to 30s total • Memory: store=True (55d paid / 1d free) & previous_interaction_id

2. SDK Installation, Model Matrix & Media Specifications

Every workflow in this guide runs on the unified late-2026 Google GenAI SDKs. Install the current packages below so typed Interactions steps, output_video helpers, and 3.8 Live fields work out of the box.

Python 3.10+ Setup (google-genai >= 2.30.1)
Prompt / Snippet
pip install -U google-genai
from google import genai
client = genai.Client()  # reads GEMINI_API_KEY
JS / TypeScript Setup (@google/genai >= 2.29.1)
Prompt / Snippet
npm install @google/genai
import { GoogleGenAI } from '@google/genai';
const ai = new GoogleGenAI({});
Strict SDK Lock: Never Install Legacy Packages
The legacy google-generativeai (Python) and @google/generative-ai (JavaScript) packages are permanently retired. They lack the Gemini 3.8 Live protocol fields and the Interactions steps schema required by Gemini Omni 1.2 Flash.

Late-2026 Live & Omni Model Matrix

Model ID Transport Primary Superpower Key Protocol Rule
gemini-3.8-live WebSocket Default low-latency voice & vision dialogue with interleaved reasoning. Async tools default; omit thinking_level.
gemini-3.8-live-extended-thinking WebSocket Deep multi-step background reasoning with spoken conversational fillers. Requires NON_BLOCKING tools & interaction_status.
gemini-3.5-transcribe-live WebSocket Real-time streaming STT with interim hypotheses & final transcripts. smart vs verbatim mode + Hybrid VAD.
gemini-3.5-live-translate-preview WebSocket Speech-to-speech live interpretation across 70+ spoken languages. Use TranslationConfig + audio transcriptions.
gemini-omni-1.2-flash Interactions API 3s–15s video creation, keyframe interpolation, edits & 30s extensions. Natural-language asset binding; ffmpeg prep.

Wire Formats & Output Ceilings

1 Live API Audio & Frame Specs
  • Audio Input: Raw little-endian 16-bit mono PCM at 16kHz native (audio/pcm;rate=16000); other rates are resampled.
  • Audio Output: Raw little-endian 16-bit mono PCM at 24kHz sample rate.
  • Video Input: JPEG frames (image/jpeg) streamed on demand via video= to save context tokens.
  • Context Budget: 128k input tokens / 64k output tokens; audio consumes ~25 tokens per second.
2 Omni 1.2 Flash Resolution & Caps
  • Aspect Ratios: Widescreen 16:9 (default) or vertical 9:16 across every quality tier.
  • Resolutions: 360p (640×360), 720p default (1280×720), 1080p (1920×1080), and 4k (3840×2160).
  • Duration Limits: 3s–15s new/first-frame/R2V (10s default); 3s–10s two-frame transitions & per-turn extensions up to 30s cumulative.
  • Prerequisites: Python 3.10+, ffmpeg, and ffprobe on PATH.

3. The Live & Omni Studio Cockpit: 7 Key Controls

Whether you run a custom web workbench or terminal pipeline, every production deployment wires together these seven live-stream and asynchronous cinema controls.

Gemini 3.8 Live + Omni 1.2 Flash Studio Cockpit
SDK v2.30.1 • wss://live & /interactions
LIVE WEBSOCKET INSPECTOR CONNECTED • 16kHz / 24kHz
OUTBOUND STREAM 1
send_realtime_input(audio=..., video=..., text=...)
Explicit keys only • Hybrid VAD audio_stream_end=True
INBOUND MULTI-PART LOOP 2
for part in server_content.model_turn.parts:
Reads ALL PCM chunks + output_transcription
BARGE-IN 3
interrupted == True
Flush speaker queue
TURN STATE 4
interaction_status
IN_PROGRESS → IDLE
OMNI 1.2 FLASH VIDEO TIMELINE 4K • 16:9 • 30s MAX
FILES API ASSET QUEUE 5
hero.png (ACTIVE) • scene_10s.mp4 (ACTIVE)
Up to 6 --image refs • Natural-language order
STATEFUL TURN CHAIN 6
previous_interaction_id="v1_9f8a..."
Turn 1 (10s) → Turn 2 Edit → Turn 3 Extend (+10s)
RENDER ARTIFACT
interaction.output_video.uri → files.download()
7

7-Point Cockpit Reference Legend

1 send_realtime_input Dispatcher: Streams live user audio (audio=), JPEG video frames (video=), and text (text=). Never pass a generic media= argument.

2 server_content Multi-Part Reader: Loops over every element inside model_turn.parts because audio bytes and metadata often arrive together.

3 interrupted Audio-Queue Flush: Fires the instant the user speaks over the assistant—immediately drop buffered 24kHz audio chunks on the client.

4 interaction_status Lifecycle Monitor: Tracks IN_PROGRESS versus IDLE so Extended Thinking clients wait through background tool calls.

5 Files API ACTIVE Upload Poller: Uploads local images or trimmed MP4s and polls until state reaches ACTIVE before calling interactions.create.

6 previous_interaction_id Turn Chain: Links iterative edits and 3s–10s scene extensions without re-uploading source video files (stored 55d paid / 1d free).

7 output_video Downloader: Reads interaction.output_video.uri and streams the finished MP4 to disk via client.files.download(file=...).

Pro Tip: Always wear headphones when testing bidirectional voice agents so speaker audio never triggers false self-interruptions.

4. Duplex Streaming & Background Extended Thinking

Gemini 3.8 Live separates real-time conversational flow from deep problem-solving. While gemini-3.8-live replies instantaneously, gemini-3.8-live-extended-thinking reasons quietly in the background and speaks natural fillers while async tools execute.

Bi-directional cerulean voice conduit above a translucent violet reasoning chamber with async tool filaments
Figure 1: Duplex 16kHz/24kHz audio flows smoothly across the top WebSocket channel while the Extended Thinking chamber runs multi-step reasoning and NON_BLOCKING tool filaments underneath without freezing speech.
1 Explicit Stream Keys & Context Injection

Connect with client.aio.live.connect (Python) or ai.live.connect (JS). Always use explicit audio=, video= (image/jpeg), and text= parameters on send_realtime_input—never pass media=.

Need to inject structured history mid-call? Call send_client_content with explicit user or model roles. Note that setting turn_complete=True unconditionally interrupts active speech.

2 Extended Thinking & interaction_status

On gemini-3.8-live-extended-thinking, pass ThinkingConfig(thinking_level="low") ("low", "medium", or "high"; minimal is unsupported). Every function declaration must set behavior="NON_BLOCKING"—blocking tools raise an API error.

Do not stop listening on turn_complete alone: monitor interaction_status ("IN_PROGRESS" vs "IDLE") and read every part inside server_content.model_turn.parts.

Prompt / Snippet
from google import genai
from google.genai import types
client = genai.Client()
lookup_crm = types.FunctionDeclaration(
    name="lookup_crm", description="Fetch customer ticket status", behavior="NON_BLOCKING",
    parameters={"type": "OBJECT", "properties": {"email": {"type": "STRING"}}, "required": ["email"]},
)
config = types.LiveConnectConfig(
    response_modalities=[types.Modality.AUDIO],
    thinking_config=types.ThinkingConfig(thinking_level="low"),
    tools=[types.Tool(function_declarations=[lookup_crm])],
)
async with client.aio.live.connect(model="gemini-3.8-live-extended-thinking", config=config) as session:
    await session.send_realtime_input(audio=types.Blob(data=pcm_16k, mime_type="audio/pcm;rate=16000"))
    await session.send_realtime_input(video=types.Blob(data=jpeg_bytes, mime_type="image/jpeg"))
    async for msg in session.receive():
        if msg.server_content and msg.server_content.model_turn:
            for part in msg.server_content.model_turn.parts:  # Always iterate ALL parts
                if part.inline_data: play_24k_pcm(part.inline_data.data)
        if getattr(msg, "interaction_status", None) == "IDLE": break

5. Live Transcribe, 70+ Language Translate & Prompt Anatomy

Beyond open-ended assistants, the Live API includes dedicated low-latency models for streaming captions and simultaneous interpretation—plus turnkey WebRTC partners for browser and telephony apps.

1 gemini-3.5-transcribe-live

Streams live speech-to-text with interim hypotheses and finalized sentences using response_modalities=["TEXT"]. Choose mode: "smart" to clean filler words and format punctuation, or "verbatim" (default) for exact word-for-word logs.

Prompt / Snippet
// Hybrid VAD: instant flush on local silence
session.sendRealtimeInput({ audioStreamEnd: true });
# Python: await session.send_realtime_input(audio_stream_end=True)
2 gemini-3.5-live-translate-preview

Translates speech directly into natural spoken audio across 70+ languages while automatically detecting the speaker's source tongue. Configure TranslationConfig inside LiveConnectConfig and enable both input and output audio transcriptions for live dual captions.

Prompt / Snippet
translation_config=types.TranslationConfig(
    target_language_code="es",
    echo_target_language=True,
)

WebRTC & Telephony Partner Bridge

Browser & Phone Edge Delivery Without Managing Raw Audio Buffers
While the core Live API speaks WebSockets from your backend, six official partner integrations bridge edge clients over WebRTC or SIP: LiveKit (Agents), Pipecat by Daily, Fishjam by Software Mansion, Vision Agents by Stream, Voximplant (inbound/outbound telephony), and the Firebase AI SDK for direct mobile and web apps.

Color-Coded Anatomy: Live Director & Omni Prompt Blueprint

Live System Persona & Modality → Thinking & Async Tool Policy → Subject & Reference Order → Timed Action Beats → Camera & Motion Style → Lighting & Atmosphere → Sound Design & Negatives
Combined Session Example:
Audio-only director persona, concise warm coaching thinking_level="low", NON_BLOCKING asset lookup • Omni Prompt: The ceramicist from the first image shapes clay in the studio from the second image [0-5s] hands center the spinning stoneware; [5-10s] she lifts the rim cleanly Continuous unbroken medium tracking shot, 35mm shallow depth of field Warm late-afternoon window rays illuminating floating clay dust Soft wheel hum and gentle ambient strings; no dialogue, no scene cuts.

6. Gemini Omni 1.2 Flash: Interactions API & Keyframes

Every gemini-omni-1.2-flash creation runs through client.interactions.create, pairing uploaded Files API assets with natural-language directions to render 3s–15s videos from scratch or between exact keyframes.

1 Files API & Storage

Upload images or MP4s via upload_file.py and poll until status is ACTIVE. Interactions default to store=True (retained 55 days paid / 1 day free) so later turns need only previous_interaction_id.

2 Text & First Frame (3s–15s)

Generate 3s–15s clips (10s default) from text alone or a single starting image (--first-frame). Choose 360p, 720p, 1080p, or 4k—pass --timeout 900 for heavy 4K renders.

3 Keyframe Transitions (3s–10s)

Supply both --first-frame and --last-frame to interpolate smoothly between start and end art (capped at 10s). Pass the same image twice to craft seamless looping videos.

Prep & Inspect Source Video with ffmpeg

Large raw camera clips slow down uploads and hit duration limits. Always check metadata with inspect_video.py (powered by ffprobe) and normalize clips with prep_video.py—defaulting to a clean 10s trim scaled to 720p/24fps without stretching, or passing --duration 3 for motion-reference clips and --duration 15 for full 15s source edits.

Prompt / Snippet
# 1. Inspect resolution, FPS, audio stream & duration; normalize to 10s 720p
./scripts/video/inspect_video.py raw_footage.mp4 --json
./scripts/video/prep_video.py raw_footage.mp4 --start 00:03 --duration 10 --output media/prepped_10s.mp4

# 2. Render 15s 4K clip with bracketed timecodes & extended HTTP timeout
./scripts/video/generate_video.py \
  "[0-5s] Runner stretches in morning mist [5-10s] Sprints along wet coastal road [10-15s] Wide aerial sunrise" \
  --duration 15 --resolution 4k --timeout 900 --output media/coastal_run_4k.mp4

# 3. Seamless 8s product loop (identical first and last frame)
./scripts/video/generate_video.py "Crystal prism spins smoothly in place" \
  --first-frame prism.png --last-frame prism.png --duration 8 --resolution 1080p --output media/loop.mp4
Single-Shot Control & Timecode Prompting
By default, Omni 1.2 Flash edits multi-shot mini-stories. Add "in a single continuous unbroken shot, no scene cuts" to lock one camera take, or use [0-5s], [5-10s], [10-15s] beats to choreograph exact transitions and audio swells.

7. Reference Binding, Surgical Edits & 30-Second Extensions

Omni 1.2 Flash abandons rigid XML markup in favor of natural-language asset binding—letting you guide characters with up to 6 reference images, refine existing clips, strip audio for new scores, and extend stories to 30 seconds.

Glass keyframe monoliths flowing into an extended magenta and cyan light reel across temporal segments
Figure 2: Keyframe monoliths and uploaded clips feed forward in 3s–10s extension hops—blending the final frames and last 10s of context up to a 30-second cumulative master.
Workflow Mode Asset & Length Caps Critical Rule for Clean Results
Natural-Language Binding Order in input[] array Refer to "the first image", "the second image", or "the reference video" in prose. Never use legacy angle-bracket placeholder tags.
Reference Images (--image) Up to 6 images (even 15s) Works across new generation, video editing, and prompt-based extension. Omit --task when extending with --image!
Reference Videos (-vr) Max 3 (3–10s) / 1 (11–15s) Strictly restricted to new video generation only (never editing or extension). Ideal reference clip length is ~3s.
Surgical Video Editing Source video ≤ 15s Keep prompts concise and end with "Keep everything else the same." Add --strip-audio to regenerate music and SFX from scratch.
Video Extensions 3s–10s/turn, 30s max Uses up to the final 10s as lookback context. Timestamps in extension prompts start at 0s for the newly added segment.
Regional Restriction Guardrail (EEA, UK, Switzerland & Select US States)
Uploading source videos for video-to-video edits, extensions, or --video-reference is not available in the EEA, United Kingdom, Switzerland, and select US states. If a video-input job returns almost immediately with zero output tokens or no video URI, check your execution region first.

8. 5 Real-World Copy-Paste Production Recipes

Drop these five proven patterns directly into your Python, TypeScript, or CLI pipelines—each tuned to late-2026 SDK signatures and safe parameter limits.

Recipe & Model Goal & Trigger Copy-Paste Snippet / Command
1. Extended Thinking Voice Support Agent
gemini-3.8-live-extended-thinking
Look up orders in CRM asynchronously while speaking warm conversational fillers, finishing only on IDLE. tool = types.FunctionDeclaration(name="get_order", behavior="NON_BLOCKING", ...)
cfg = types.LiveConnectConfig(response_modalities=["AUDIO"], thinking_config=types.ThinkingConfig(thinking_level="medium"), tools=[types.Tool(function_declarations=[tool])])
Send tool output via session.send_tool_response(...); wait for interaction_status == "IDLE".
2. Hybrid VAD Bilingual Translator
gemini-3.5-live-translate-preview
Simultaneous speech-to-speech interpretation across 70+ languages with instant turn flush on mic pause. cfg = types.LiveConnectConfig(response_modalities=["AUDIO"], translation_config=types.TranslationConfig(target_language_code="ja", echo_target_language=True), input_audio_transcription=types.AudioTranscriptionConfig(), output_audio_transcription=types.AudioTranscriptionConfig())
On client silence: await session.send_realtime_input(audio_stream_end=True).
3. Two-Frame Loop → 1080p Timeline
gemini-omni-1.2-flash
Create an 8s looping hero product clip, then extend it into a 1080p multi-scene showcase with new talent. ./scripts/video/generate_video.py "Smooth 360 orbit" --first-frame watch.png --last-frame watch.png --duration 8 --resolution 1080p --output media/watch_loop.mp4
./scripts/video/generate_video.py "Extend this video: [0-5s] model from the reference image lifts the watch [5-10s] macro dial gleam" --extend media/watch_loop.mp4 --image model.png --duration 10 --resolution 1080p --output media/watch_story.mp4
4. Stateful Wardrobe Swap + Score Regen
gemini-omni-1.2-flash
Swap clothing via previous_interaction_id, then strip audio on a local clip to compose a brand-new soundtrack. res2 = client.interactions.create(model="gemini-omni-1.2-flash", previous_interaction_id=res1.id, input=[{"type":"image","uri":coat_uri,"mime_type":"image/png"}, {"type":"text","text":"Replace the jacket with the crimson coat from the reference image. Keep everything else the same."}])
./scripts/video/generate_video.py "Transform lighting to neon noir; upbeat synthwave score" --video media/scene.mp4 --strip-audio --output media/noir.mp4
5. Parallel Omni Batch + Files Uploader
gemini-omni-1.2-flash
Pre-upload assets to ACTIVE and execute mixed generation, interpolation, and editing jobs concurrently. ./scripts/upload_file.py hero.png → wait for ACTIVE URI
./scripts/video/generate_video.py --batch jobs.json --concurrency 3
Each entry in jobs.json declares prompt, duration, resolution, optional first_frame/last_frame/image/video_reference, and output.
Why These 5 Recipes Never Fail in Production
Notice how Recipe 1 pairs behavior="NON_BLOCKING" with interaction_status, Recipe 3 omits --task so --image works during extension, and Recipe 4 uses --strip-audio so Omni generates a fresh score instead of preserving the original track.

9. Production Session Hardening, Tokens & Regional Guardrails

Demo scripts run for 30 seconds; production apps stay connected for hours from untrusted browsers. Use these four guardrails to eliminate dropped calls, context limits, and leaked API keys.

1 Single Modality & Compression Caps

Each Live session allows only one response modality: AUDIO or TEXT, never both. Need text alongside native voice? Request response_modalities=["AUDIO"] and enable output_audio_transcription.

Uncompressed sessions cap at 15 minutes (audio-only) or 2 minutes (audio + video). Enable ContextWindowCompressionConfig(sliding_window=SlidingWindow()) to run indefinitely.

2 10m WebSockets, Resumption & GoAway

Individual WebSocket connections reset roughly every 10 minutes. Enable SessionResumptionConfig so the server streams session_resumption_update handles (valid for 2 hours) to reconnect seamlessly without losing memory.

Watch for response.go_away (inspecting time_left) so your client opens a fresh resumed socket before the old link closes.

Client-Side Security: Ephemeral Auth Tokens

Never embed your master GEMINI_API_KEY in browser or mobile code. Mint short-lived ephemeral tokens on your backend (default new_session_expire_time 1 minute, expire_time 30 minutes) and lock model/config constraints server-side using live_connect_constraints.

Prompt / Snippet
# Backend (Python): mint locked single-use ephemeral token for v1beta Live WebSocket
token = client.auth_tokens.create(config={
    "uses": 1,
    "live_connect_constraints": {
        "model": "gemini-3.8-live",
        "config": {"response_modalities": ["AUDIO"], "session_resumption": {}, "context_window_compression": {"sliding_window": {}}},
    },
})
// Frontend (JS): pass token.name as apiKey into @google/genai
const ai = new GoogleGenAI({ apiKey: token.name });
Regional Video Upload Guardrail & Stateless Storage Opt-Out
EEA / UK / Switzerland / Select US States: Text-to-video and image-to-video work globally, but uploading videos for editing, extension, or --video-reference is blocked regionally—surface a clear UI fallback before users upload large clips. For zero-retention compliance on Omni, passing store=False disables server history (and turns off previous_interaction_id and background=True).

10. End-to-End Studio: Live Voice Director to 4K Omni Master

Combine Gemini's multimodal lineup into a single automated production house—from live spoken creative direction and bespoke Nano Banana 2.1 keyframes to 4K Omni cinema, expressive TTS voiceover, and WebRTC preview.

STAGE 1 • DIRECT Live Voice + Concept • gemini-3.8-live • Spoken creative brief • NON_BLOCKING trigger gemini-nano-banana-2.1 • Generates start/end keyframes & style refs Output: Beat sheet + PNGs STAGE 2 • PREP ffmpeg & Files API • inspect_video.py • prep_video.py (10s/3s) • Optional --strip-audio upload_file.py • Uploads assets & polls until state == ACTIVE Output: Normalized URIs STAGE 3 • RENDER Omni 1.2 Flash 4K • gemini-omni-1.2-flash • Keyframes + up to 6 refs • Surgical edits (keep same) previous_interaction_id • Chain 3s–10s extensions up to 30s 4K (--timeout 900) Output: 30s 4K MP4 Master STAGE 4 • DELIVER Voiceover & Review • gemini-3.8-flash-tts • Multi-speaker narration • Voice Design personas WebRTC & Live Review • LiveKit / Pipecat / Firebase • Client screenshare QA Output: Ready to Publish
1 Voice-Driven Storyboarding

Talk through scene ideas with gemini-3.8-live. When you approve a shot list, an async tool invokes gemini-nano-banana-2.1 to render consistent character sheets and start/end keyframes.

2 Automated Omni Assembly

Feed those Nano Banana 2.1 keyframes into generate_video.py --batch jobs.json, normalizing any live-action B-roll through prep_video.py before chaining 30-second 4K extensions.

3 Expressive TTS & Live Review

Layer studio narration from gemini-3.8-flash-tts, then stream the preview inside a LiveKit or Firebase WebRTC session so stakeholders critique frames by voice in real time.

11. Troubleshooting Clinic (Top 7 Fixes) & Pre-Flight Checklist

Bookmark this final page. Almost every runtime error in Live streaming or Omni video creation traces back to one of these seven protocol mismatches.

Symptom / Error Root Cause 10-Second Fix
1. media= Rejected in send_realtime_input Legacy generic media key is unsupported on 3.8 Live. Replace media= with explicit audio=, video= (image/jpeg), or text= parameters.
2. Hard Error on Tool Call in Extended Thinking Synchronous blocking functions fail on gemini-3.8-live-extended-thinking. Set behavior="NON_BLOCKING" on every FunctionDeclaration and track interaction_status.
3. Missing Audio Chunks or Silent Gaps Client reads only parts[0] of server_content.model_turn. Iterate over all items in model_turn.parts—audio bytes and metadata arrive in multi-part events.
4. --last-frame Fails or Rejects >10s End frame passed without start frame, or duration set to 11s–15s. Always pair --last-frame with --first-frame and keep keyframe transitions at 3s–10s max.
5. --image Blocked During Extension Explicitly passing --task extend forces a single-video gateway lock. Omit --task and use natural-language extension ("Extend this video...") so up to 6 --image refs work.
6. --video-reference Rejected on Job Used on edit/extend, or >1 ref video on an 11s–15s clip. Use --video-reference only on new generation (max 3 for 3–10s, max 1 for 11–15s; ~3s ideal).
7. Omni Video Edit Finishes Fast & Empty Zero output tokens from EEA, UK, Switzerland, or restricted US state. Video uploads for edits, extensions, and R2V are regionally blocked there; run from a supported region.

6-Point Production Pre-Flight Checklist

  • SDK & Binary Lock: Confirm google-genai >= 2.30.1 or @google/genai >= 2.29.1, Python 3.10+, and ffmpeg / ffprobe on PATH.
  • Audio & Barge-In Ready: Stream 16kHz mono PCM in, play 24kHz PCM out, flush queue on interrupted=True, and fire audio_stream_end on pause.
  • Long-Session Resilience: Enable context_window_compression, persist session_resumption handles (2h TTL), and handle GoAway signals.
  • Browser Security: Mint server-side ephemeral tokens with live_connect_constraints; never ship GEMINI_API_KEY to client bundles.
  • Omni Asset Hygiene: Trim source videos via prep_video.py, poll Files API to ACTIVE, bind assets by natural-language order, and set --timeout 900 for 4K.
  • Audio Choice on Edits: Leave original track intact by default, or pass --strip-audio to force full score and SFX regeneration from scratch.

Save the 12-Page Magazine PDF

Keep a print-ready copy of Gemini 3.8 Live & Omni 1.2 Flash Studio for offline reading, desk reference, or sharing with teammates (3.8 MB).