Models

Every AI model in OpenCut AI is open source and runs locally on your machine. Cloud APIs available for Indian and multilingual content.

Speech-to-TextLanguage ModelImage GenerationText-to-SpeechIndian LanguagesMultilingual TTS/STTSpeaker DetectionFace DetectionEmotion DetectionModel OptimizationBackground RemovalAudio Processing

Performance Tiers

Lite

4-8GB

Llama 3.2 1B + Whisper Base + noisereduce

Runs on any machine. Basic AI commands, transcription, and audio cleanup. TurboQuant lets you squeeze a 3B model in here.

Recommended

Standard

8-16GB

Llama 3.2 3B / Mistral 7B + Whisper Small + XTTS v2 + rembg

Recommended for most users. Good AI commands, voice cloning, and image generation. TurboQuant compression makes this the sweet spot.

Pro

16-32+GB

Llama 3.1 8B + Whisper Medium + SDXL + XTTS v2 + full stack

Best quality across the board. 8B models with near-lossless 4-bit compression. GPU recommended for image generation.

TurboQuant Memory Compression

Built-in

All tiers benefit from TurboQuant KV cache compression. Full AI stack memory dropped from 35 GB to 15 GB. One-click setup in Settings auto-detects your hardware and recommends the best configuration.

4-bit

0.9986 cosine sim

Near-lossless

3-bit

0.9953 cosine sim

6x compression

2-bit

0.9874 cosine sim

7.3x compression

All models

Whisper

OpenAI

Speech-to-Text

Automatic speech recognition with word-level timestamps. Powers transcription, subtitles, and text-based editing. Available from tiny (39M) to large-v3 (1.5B). TurboQuant compression lets Whisper Medium fit where only Base could before.

transcriptionsubtitlestimestamps
39M - 1.5BMIT

Llama 3.2 / 3.1

Meta

Language Model

Powers AI commands, script writing, chapter analysis, clip finding, smart suggestions, and prompt enhancement. Runs via Ollama. Available at 1B (Lite), 3B (Standard), and 8B (Pro) with Q3/Q4/Q5 quantization.

commandsscriptsanalysischapters
1B - 8BLlama 3 Community

Mistral 7B

Mistral AI

Language Model

High quality 7B instruction-tuned model. Available at Q4 (Standard tier) and Q5 (Pro tier). Strong at structured JSON generation for editor commands and content analysis.

commandsscriptsjsonanalysis
7BApache 2.0

Qwen2.5

Alibaba

Language Model

TurboQuant-validated model family from 0.5B to 14B. The 3B Instruct variant is benchmarked with full compression data (0.9986 cosine similarity at 4-bit). Also includes Coder variants for technical content.

turboquantcommandscodinganalysis
0.5B - 14BApache 2.0

Stable Diffusion XL

Stability AI

Image Generation

Generate images from text prompts for video overlays, thumbnails, and B-roll. Supports SDXL Turbo (4-step fast generation) and full SDXL (20-step quality).

imagesoverlaysthumbnails
3.5B - 6.6BOpenRAIL-M

FLUX.1

Black Forest Labs

Image Generation

Next-generation text-to-image with superior prompt adherence and image quality. Alternative to SDXL for higher quality results.

imagesoverlayshigh-quality
12BApache 2.0

XTTS v2

Coqui

Text-to-Speech

Local voice synthesis with voice cloning from a 6-second sample. Supports 17+ languages. TurboQuant at 3-bit shrinks it from 1.8 GB to 0.6 GB while preserving voice quality.

voiceovervoice-cloninglocal
467MMPL 2.0

Sarvam AI

Sarvam AI

Indian Languages

Purpose-built for 22 Indian regional languages. Saaras v3 for transcription, Bulbul v3 for text-to-speech with 37+ natural voices, plus translation and transliteration.

indian-languagestranscriptionttstranslation
Cloud APIAPI Key

Smallest AI

Smallest AI

Multilingual TTS/STT

Ultra-fast cloud TTS and STT. Lightning TTS with 80+ voices across 15 languages at ~100ms latency. Pulse STT covers 39 languages with speaker diarization and emotion detection.

ttssttmultilingualfast
Cloud APIAPI Key

Pyannote

Herve Bredin

Speaker Detection

Neural speaker diarization that identifies who is talking when. Powers multi-speaker detection and automatic speaker-boundary cuts. Falls back to FFmpeg silence-based detection.

speakersdiarizationpodcast
~90MMIT

MediaPipe

Google

Face Detection

Real-time face detection for auto-reframe. Tracks faces across video frames to generate 9:16 crops for TikTok, Reels, and Shorts.

face-trackingauto-reframe9:16
~1MApache 2.0

SpeechBrain

SpeechBrain

Emotion Detection

Detects emotional peaks in audio for finding the most impactful moments. Used for clip scoring and highlight detection. Falls back to energy-based analysis locally.

emotionshighlightsintensity
~300MApache 2.0

TurboQuant

OpenCut AI

Model Optimization

KV cache compression using Google Research's PolarQuant + QJL. Achieves 6x memory reduction at 3-bit with 0.9953 cosine similarity. Makes 7B models run on 8 GB RAM. One-click setup in Settings.

compressionmemoryoptimizationkv-cache
CompressionMIT

U-Net (rembg)

danielgatis

Background Removal

Remove backgrounds from images to create transparent overlays. Useful for speaker cutouts, product shots, and compositing.

bg-removalcompositing
176MMIT

noisereduce

Tim Sainburg

Audio Processing

Spectral gating noise reduction for cleaning up audio recordings. Removes background noise, hiss, and hum with adjustable strength.

audiodenoisingcleanup
< 1MMIT