AI models

Jev

Jev is TypeSafe AI's hosted System One model for making typed decisions inside software. Give it text or structured text state and ask narrow questions of three kinds: a Choice picks one option from a set you define, a Score rates the input against levels you describe, and a Noul is a yes-or-no question. Each answer comes with probabilities your code can use for routing, classification and scoring. It does not generate prose or code and cannot replace a chat or coding-agent LLM. The documented Jev 1.13 version is available through TypeSafe's API in early access. Start with a decision in your own workflow and keep the final action in your application code.

Cloud · Paid

Ornith-1.5-35B-A3B

Ornith-1.5-35B-A3B is an open-weight reasoning model for coding, repository work and agents that call tools. Its Qwen3.5-family mixture-of-experts architecture contains about 35 billion parameters but activates roughly 3 billion per token. Ornith describes training that jointly improves task generation, agent scaffolds and solutions through reinforcement learning. The published context window is 262,144 tokens. It is a candidate for developers who want to run a coding assistant on their own infrastructure: the official card documents vLLM and SGLang servers with OpenAI-compatible APIs, plus GGUF use through Ollama or llama.cpp. Start with a small repository task and check the result against your tests before expanding its role.

Self-hosted · Free

MiniCPM5-2B

A compact OpenBMB text model with downloadable weights, tool calling and several local runtime formats.

Self-hosted · Desktop · Free

XTTS-v2

Synthesize speech in multiple languages from text using a voice reference.

Self-hosted · Free

Whisper (open source)

Transcribe lectures, interviews, and recordings with an openly released speech model.

Self-hosted · Free

Wan 2.2

Generate short video clips from text or animate an existing image.

Self-hosted · Free

Google Veo

Generate video scenes from text prompts using Google's cloud model.

Cloud · Paid

Sora 2 (discontinued)

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.

Cloud · Paid

SadTalker

Animate a still image into a talking head from an audio clip.

Self-hosted · Free

Qwen3.6 (35B-A3B)

Run an open mixture-of-experts model with coding and multilingual capabilities.

Self-hosted · Free

Phi-4-mini

Run a compact local model for summaries, drafts, classification, and simple code.

Self-hosted · Free

OpenAI TTS

Turn written content into spoken narration with OpenAI TTS models.

Cloud · Bring your own key

OpenAI Realtime API

Stream microphone audio and receive spoken responses in one API.

Cloud · Paid

Nano Banana Pro (Gemini 3 Pro Image)

Generate images with text, data labels, and multimodal instructions.

Cloud · Paid

Moshi (Kyutai)

Build speech-native conversations that listen and speak simultaneously.

Self-hosted · Free

Ministral 3

Run small open-weight models on laptops or edge hardware.

Self-hosted · Free

Mistral Medium 3.5

Process text and images with a large-context multimodal model.

Cloud · Paid

MiniMax-M2

Use a model focused on coding and agentic tool chains.

Self-hosted · Paid

LTX-Video

Generate video from text or images through ComfyUI workflows.

Self-hosted · Free

Llama 4 Scout

Run a multimodal model for broad tasks with a large context window.

Self-hosted · Free

LatentSync

Align supplied audio with mouth motion in existing face video.

Self-hosted · Free

Kokoro 82M (open source)

Generate speech locally with a lightweight Apache-2.0 model.

Self-hosted · Free

Kling

Generate high-resolution video with controlled motion.

Cloud · Paid

HunyuanVideo

Generate short text-to-video clips with an open model.

Self-hosted · Free