Synthesia
View profileWeb servicesCloud
CommercialWhat you can do
- Turn a script or PowerPoint deck into a presenter-led training video.
Before choosing
- Its corporate training workflow is excessive for a single quick video.
Tools
Find tools for your task. Compare apps, models, providers and runtimes.
180tools
Sora 2 (discontinued)
View profileAI modelsCloud
CommercialWhat you can do
- Discontinued. OpenAI closed the Sora web and app experiences on 26 April 2026 and the Sora API on 24 September 2026. Export any content you still need; use Veo, Runway or Kling for new video work.
Before choosing
- Discontinued: don't start new work on Sora. Export existing content at sora.chatgpt.com/sunset before OpenAI deletes it.
Stable Diffusion 3.5 + ComfyUI
View profileLocal & self-hosted appsSelf-hosted
FreeWhat you can do
- Build repeatable image workflows by connecting generation, ControlNet, LoRA, and upscaling nodes.
Before choosing
- The license has revenue conditions, and ComfyUI requires learning its node graph.
Scaleway
View profileModel providersCloud
CommercialWhat you can do
- Call open models through an OpenAI-compatible API in French data centers.
Before choosing
- Generative APIs and GPU Instances require different operating responsibilities.
SadTalker
View profileAI modelsSelf-hosted
FreeWhat you can do
- Animate a still image into a talking head from an audio clip.
Before choosing
- Self-hosting requires setup and a capable GPU for usable speed.
Runway Gen-4
View profileWeb servicesCloud
CommercialWhat you can do
- Direct camera motion and preserve subjects across generated video shots.
Before choosing
- Its production-oriented controls add complexity for a single illustrative clip.
Retell AI
View profileWeb servicesCloud
BYOKWhat you can do
- Assemble a phone-ready voice agent from your chosen speech and language services.
Before choosing
- You must provide separate speech-to-text, language-model, and speech keys.
Requesty
View profileModel providersCloud
Free tierWhat you can do
- Route requests to many language-model providers through one compatible API.
Before choosing
- The gateway adds a markup to the underlying provider model fees.
Qwen3.6 (35B-A3B)
View profileAI modelsSelf-hosted
FreeWhat you can do
- Run an open mixture-of-experts model with coding and multilingual capabilities.
Before choosing
- Its model size still requires substantial memory for local inference.
Promptfoo
View profileDevelopment & infrastructureSelf-hosted · Cloud
Free tierOpen sourceWhat you can do
- Define repeatable evaluations for prompts, models, providers, and datasets.
Before choosing
- Results are only as useful as the test cases and assertions you define.
Poe
View profileWeb servicesCloud
CommercialWhat you can do
- Chat with official model bots and user-built bots from one interface.
Before choosing
- Third-party bots can handle data differently, so choose bots carefully.
Podman
View profileDevelopment & infrastructureDesktop · Self-hosted
FreeOpen sourceWhat you can do
- Run OCI containers without a central daemon.
Before choosing
- podman compose depends on an external provider whose behavior varies by environment.
Pipecat
View profileDevelopment & infrastructureSelf-hosted
FreeWhat you can do
- Assemble voice activity detection, transcription, language, and speech components.
Before choosing
- You must integrate and self-host the separate pipeline components.
Phi-4-mini
View profileAI modelsSelf-hosted
FreeWhat you can do
- Run a compact local model for summaries, drafts, classification, and simple code.
Before choosing
- Its small size limits performance on difficult reasoning and niche knowledge tasks.
Perplexity Max
View profileWeb servicesCloud
CommercialWhat you can do
- Access several models and Perplexity search features through one subscription.
Before choosing
- Reported Model Council and Background Assistants details still need primary confirmation.
Perplexity
View profileWeb servicesCloud
Free tierWhat you can do
- Ask natural-language questions and receive web answers with clickable citations.
Before choosing
- Check every citation and quotation against the original source.
OVHcloud
View profileModel providersCloud
CommercialWhat you can do
- Call open-weight models through a serverless inference API in European data centers.
Before choosing
- Its AI-specific features may be narrower than those of dedicated inference specialists.
OpenVINO GenAI
View profileDevelopment & infrastructureSelf-hosted
FreeWhat you can do
- Run language, vision, image, speech-to-text, and text-to-speech models locally.
Before choosing
- Choose compatible models and hardware before building an OpenVINO workflow.
OpenRouter
View profileModel providersCloud
Free tierWhat you can do
- Connect an application to hosted models through one API, with provider routing, fallbacks and usage accounting.
Before choosing
- Model prices and provider data policies differ. Credit purchases add fees; free variants have limits. Default routing is not a retention-restricted configuration.
OpenHands
View profileWeb servicesLocal & self-hosted appsCloud · Self-hosted
Free tierOpen sourceWhat you can do
- Run an autonomous coding agent that plans and applies repository changes.
Before choosing
- Use a sandbox because an unsandboxed agent server can access your filesystem.
opencode + voice
View profileLocal & self-hosted appsSelf-hosted
FreeWhat you can do
- Drive opencode hands-free by speaking commands into a local voice pipeline.
Before choosing
- This combines separately maintained community projects rather than an official opencode feature.
opencode
View profileLocal & self-hosted appsSelf-hosted
BYOKOpen sourceWhat you can do
- Use a terminal, desktop app, or IDE extension to work with coding agents.
Before choosing
- Bring your own provider access and monitor its token usage.
openclaw
View profileLocal & self-hosted appsSelf-hosted
Free tierWhat you can do
- Run a personal agent that you reach through familiar messaging apps.
Before choosing
- Setup needs Node 24+, an LLM API key, and a local host machine.
OpenAI TTS
View profileAI modelsCloud
BYOKWhat you can do
- Turn written content into spoken narration with OpenAI TTS models.
Before choosing
- The available voices may not suit projects needing cloning or especially realistic delivery.