Veo 3.1 vs Kling 3.0 vs Runway vs MiniMax H3: Best AI Video Generator After Sora

Q: Which AI video generator should I use?

Short social clips with sound: Veo 3.1 Fast or Lite Story scenes with talking characters: Kling 3.0 Editing video you already filmed: Runway Aleph Privacy, control or high volume on your own hardware: MiniMax H3 Marketing teams on a budget: compare our AI video pricing guide

Comparisons

Google Veo 3.1, Kling 3.0, Runway Gen-4.5 and open-source MiniMax H3 compared on clip length, audio, price per second and best use, now that Sora is gone.

AP

Staff

November 10, 2026 · 2 min read

With OpenAI’s Sora shut down, the best AI video generators in late 2026 are Google Veo 3.1 for sound and polish, Kling 3.0 for longer clips with dialogue at a low price, Runway Gen-4.5 for editing real footage, and MiniMax H3 if you want open weights you can run yourself. Here is how they compare, in plain English.

What happened to Sora?

OpenAI closed the Sora app and website on April 26, 2026, and the API on September 24. Reports said it cost far more to run than it earned, and OpenAI moved the computing power to coding and enterprise work. Our Sora shutdown timeline has the full story.

How do Veo, Kling, Runway and MiniMax compare?

Key facts from an October roundup by OCDevel and Digital Applied; prices are approximate API rates per second of video and change often:

Google Veo 3.1: clips of 4, 6 or 8 seconds, extendable in 7-second steps; native sound, effects and dialogue; three speed tiers from about 0.05 dollars a second (Lite) to about 0.40 (Quality). In the Gemini app it has been replaced by Gemini Omni, which allows chat-style video editing
Kling 3.0 (Kuaishou): up to 15 seconds per shot, multi-shot scenes with consistent characters, lip-synced dialogue in five languages; about 0.08 dollars a second silent and 0.13 with audio on fal
Runway Gen-4.5: about 0.12 dollars a second, no native dialogue; its strength is Aleph, which relights, re-angles and edits existing footage, plus Act-Two for moving a filmed performance onto a character
MiniMax H3: open weights, 33 billion parameters, stereo audio, up to 2K and 15 seconds; free to run if you have about four GPUs, with a commercial licence that some countries must apply for

Which AI video generator should I use?

Short social clips with sound: Veo 3.1 Fast or Lite
Story scenes with talking characters: Kling 3.0
Editing video you already filmed: Runway Aleph
Privacy, control or high volume on your own hardware: MiniMax H3
Marketing teams on a budget: compare our AI video pricing guide

Which one has the best quality?

It depends on the clip. On blind-vote video leaderboards the top spots change every few weeks, and a hosted, fine-tuned version of MiniMax H3 has ranked among the top five. Run the same prompt through two or three tools before committing, since most offer free credits.

Are AI videos labelled?

Google stamps all Veo and Omni output with an invisible SynthID watermark. Others vary. With realistic AI video now cheap, read our guide to AI video and deepfakes before publishing anything showing real people.

ElevenLabs Realtime v1: One Voice Model That Replaces the Whole Voice-Agent Stack

Enterprise Agents

ElevenLabs Eleven Realtime v1 Mini and Max turn a five-part voice agent stack into one model, with 0.57-second replies, 90+ languages and OpenAI Realtime API compatibility.

AP

Staff

November 10, 2026 · 2 min read

ElevenLabs has launched Eleven Realtime v1, a voice model that listens to speech and talks back directly, replacing the five or so separate pieces developers used to wire together to build a voice agent. It comes in two sizes, Mini and Max, starts speaking in about 0.6 seconds, works in more than 90 languages and is compatible with OpenAI’s Realtime API, so switching can take just a couple of code changes.

What problem does Eleven Realtime solve?

Building a phone or web voice agent used to mean stacking several systems: one to turn speech into text, one to detect when you stop talking, one to manage turn-taking, a language model to decide what to say, and a text-to-speech voice to say it. Each handoff adds delay and another thing to tune. ElevenLabs put the whole stack behind one model ID. As the company put it, you “shouldn’t have to become a configuration expert to build a voice agent that sounds natural,” per StartupHub.ai.

What is the difference between Mini and Max?

v1 Max: for complex jobs like customer support and lead qualification. About 0.57 seconds to first audio, 98.8% on the Big Bench Audio reasoning test and 87.1% success on tool-using tasks
v1 Mini: for natural everyday conversation. About 0.62 seconds to first audio, slightly better at handling pauses and turn-taking, but much weaker at reasoning (75.9%) and tools (60.3%)

ElevenLabs says these scores were evaluated by Artificial Analysis. They are company-reported figures, not our own tests.

Does it keep a text record of conversations?

Yes, and that matters for businesses. Instead of a single sealed audio-to-audio system, Eleven Realtime keeps text at every step inside its pipeline. That lets teams see exactly what the agent heard and what it decided to say, and lets ElevenLabs apply guardrails and hide personal data, which is harder with fully fused voice models.

Why does OpenAI compatibility matter?

Many voice apps are built on OpenAI’s Realtime API, such as those covered in our look at OpenAI’s GPT Live 1 for developers. Because ElevenLabs matches its message format, developers can change the endpoint and model name and keep the rest of their code. That makes it easy to test both and pick on price, voice quality or speed.

How much does it cost?

ElevenLabs charges per minute of connected audio plus language model tokens; exact rates are on its pricing page. It works inside ElevenAgents, with phone, web and WhatsApp channels, or directly through the Realtime API over WebSocket. Apps using WebRTC have to switch to WebSocket first.

Who should use it?

Teams building customer support lines, booking agents or sales assistants that need to sound natural and use tools. Any of ElevenLabs’ 17,000+ library voices can be used. Just remember the voice-cloning ground rules in our voice cloning explainer.

Loading more stories…

You've reached the end — that's every story we've published.