Доклад " Bugg " / " Особенности запроса "

AI Models We Use - All Open Source | Free.ai

Every AI model on Free.ai — all MIT/Apache 2.0 licensed. Qwen, FLUX, Kokoro, Whisper, and more.

62 models · 100% open source

Chat (9) Code (3) Image Generation (8) Video Generation (6) Text-to-Speech (9) Speech-to-Text (3) Music Generation (2) Translation (2) Vision (3) OCR (4) Image Enhancement (5) Audio Separation (2) 3D Generation (3) Embeddings (3)

Chat & Conversation 9

DeepSeek V3

MIT 685B MoE Самоорганизующиеся

DeepSeek V3 is a Mixture-of-Experts model with 685B total / 37B active parameters. MIT …

DeepSeek-R1 7B

MIT 7B Самоорганизующиеся

Reasoning-focused model that shows its thinking process. Excellent at math, logic, and complex analysis.

IBM Granite 3.3 8B

Apache 2.0 8B Самоорганизующиеся

IBM's production-grade 8B. Apache 2.0. Extensively RLHF-tuned on enterprise tasks with strict IBM AI …

Mistral 7B

Apache 2.0 7B Самоорганизующиеся

High-quality 7B model from Mistral AI with excellent instruction following.

Mistral Small 3 (24B)

Apache 2.0 24B Самоорганизующиеся

Mistral Small 3 — Apache 2.0. 24B dense, fast inference, strong multilingual. Drop-in replacement …

Phi-4 14B

MIT 14B Самоорганизующиеся

Phi-4 — MIT licensed. Microsoft's strongest small model, trained on curated synthetic data.

Qwen3 32B

Apache 2.0 32B Самоорганизующиеся

Mid-size Qwen3. Apache 2.0. Strong multilingual + reasoning. Fits on a single 20GB A100.

Qwen3 Next 80B

Apache 2.0 80B Самоорганизующиеся

Qwen3-Next-80B is Alibaba's flagship open reasoning model released late 2026. Apache 2.0 licensed. Competitive …

SmolLM 3 3B

Apache 2.0 3B Самоорганизующиеся

SmolLM 3 — Apache 2.0, Hugging Face's tiny-but-capable open model. Edge deployment friendly.

Explore more chat models

380+ external options via API

Code Generation 3

Qwen 2.5 Coder 32B

Apache 2.0 32B Самоорганизующиеся

Best open-source code model. Excels at Python, JavaScript, TypeScript, and 50+ languages.

by Alibaba Cloud

Qwen3-Coder 7B

Apache 2.0 7B Самоорганизующиеся

Dedicated code model optimized for programming tasks. Supports 100+ languages.

by Alibaba Cloud

StarCoder 2 15B

BigCode OpenRAIL-M 15B Самоорганизующиеся

StarCoder2 — OpenRAIL-M (commercial-allowed with use-restrictions). Trained on permissively-licensed GitHub.

Explore more code models

380+ external options via API

Image Generation 8

FLUX.1 Krea

Apache 2.0 12B Самоорганизующиеся

FLUX.1 Krea — Apache 2.0 open variant of FLUX tuned for photorealistic output.

by Black Forest Labs

FLUX.1 Schnell

Apache 2.0 12B Самоорганизующиеся

State-of-the-art text-to-image model. Fast generation with stunning quality.

by Black Forest Labs

FLUX.2 Klein 4B

Apache 2.0 4B Самоорганизующиеся

Latest FLUX image generation model with improved quality and style diversity.

by Black Forest Labs

Kolors

Apache 2.0 2.5B Самоорганизующиеся

Kolors — Apache 2.0. ChatGLM3-based text encoder; exceptional Chinese-English bilingual prompts.

Lumina Image 2

Apache 2.0 2.6B Самоорганизующиеся

Lumina Image 2 — Apache 2.0. DiT-based, strong prompt following, highly controllable.

PixArt-Σ

OpenRAIL++ 600M Самоорганизующиеся

PixArt-Σ — OpenRAIL++. Transformer-based 4K-capable image gen at a fraction of SDXL's compute.

by PixArt-alpha

SDXL Turbo

OpenRAIL++ 3.5B Самоорганизующиеся

SDXL Turbo — OpenRAIL++. Single-step image generation; order-of-magnitude faster than SDXL.

Stable Diffusion XL

OpenRAIL++ 3.5B Самоорганизующиеся

Industry-standard image generation model with wide community support.

by Stability AI

Explore more image models

380+ external options via API

Video Generation 6

AnimateDiff v3

Apache 2.0 1B adapter Самоорганизующиеся

AnimateDiff — Apache 2.0. Motion adapter that animates SDXL/SD 1.5 images.

CogVideoX-2B

Apache 2.0 2B Самоорганизующиеся

Text-to-video generation model producing 4-8 second clips.

by Tsinghua University

HunyuanVideo 1.5

Apache 2.0 1.5B Самоорганизующиеся

Generate stunning videos from text prompts. Higher quality than CogVideoX with more natural motion.

Mochi 1

Apache 2.0 10B Самоорганизующиеся

Mochi 1 — Apache 2.0! The only fully-open frontier text-to-video model. Strong motion, cinematic.

SadTalker

MIT 500M Самоорганизующиеся

SadTalker — MIT. Audio-driven talking-head video generation. Photo + audio in, lip-synced video out. …

Wan 2.1

Apache 2.0 14B Самоорганизующиеся

Wan 2.1 — Apache 2.0! First fully-open competitive T2V+I2V model.

Explore more video models

380+ external options via API

Text-to-Speech (TTS) 9

Bark

MIT 1.5B Самоорганизующиеся

Bark — MIT licensed text-to-audio model that generates speech, music, sound effects, and non-speech …

Chatterbox

MIT Самоорганизующиеся

Highly expressive TTS with voice cloning support.

Kokoro

Apache 2.0 Самоорганизующиеся

Ultra-fast TTS with natural intonation. Supports streaming.

MeloTTS

MIT Самоорганизующиеся

High-quality multilingual text-to-speech model.

OuteTTS

Apache 2.0 Самоорганизующиеся

TTS with built-in voice cloning from short audio samples.

Parler-TTS Large

Apache 2.0 2.3B Самоорганизующиеся

Parler-TTS — Apache 2.0. Fully open reproduction of Google's Parler. Control tone, pace, and …

Piper

MIT Самоорганизующиеся

Fast and lightweight TTS with dozens of voices and languages.

Qwen3-TTS 0.6B

Apache 2.0 0.6B Самоорганизующиеся

Advanced TTS with 3-second voice cloning. Clone any voice from a short sample. Multilingual …

by Alibaba Cloud

StyleTTS 2

MIT 24M Самоорганизующиеся

StyleTTS 2 — MIT licensed, zero-shot voice cloning from a single reference sample. Small, …

by Yinghao Aaron Li

Explore more TTS models

380+ external options via API

Speech-to-Text (STT) 3

Distil-Whisper large-v3

MIT 756M Самоорганизующиеся

Distil-Whisper — MIT licensed. 6× faster than Whisper large-v3 with ~1% WER gap. Drop-in …

Moonshine Base

MIT 61M Самоорганизующиеся

Moonshine — MIT. Optimized for edge devices; faster than Whisper tiny at better accuracy.

by Useful Sensors

faster-whisper large-v3

MIT Самоорганизующиеся

CTranslate2 port of OpenAI Whisper. Fast and accurate speech recognition in 99 languages.

by OpenAI / SYSTRAN

Explore more STT models

380+ external options via API

Music Generation 2

ACE-Step 3.5B

MIT 3.5B Самоорганизующиеся

Generate complete songs with vocals, lyrics, and instrumentation in seconds. Supports any genre and …

AudioLDM 2

Apache 2.0 Самоорганизующиеся

Generate music from text descriptions. Any genre, mood, or style.

Explore more music models

380+ external options via API

Translation 2

MadLAD-400 3B

Apache 2.0 3B Самоорганизующиеся

Google's Massive Multilingual Adaptive Dataset translation model. Supports 450+ languages with Apache 2.0 license.

Opus-MT

Apache 2.0 77M/pair Самоорганизующиеся

Opus-MT — Apache 2.0. 1000+ language-pair models. Tiny footprint, no licensing risk.

by Helsinki-NLP

Explore more translation models

380+ external options via API

Vision & Multimodal 3

InternVL 3 8B

MIT 8B Самоорганизующиеся

InternVL 3 — MIT. One of the strongest open VLMs, near Qwen-VL 72B quality …

by Shanghai AI Lab

Molmo 7B

Apache 2.0 7B Самоорганизующиеся

Molmo — Apache 2.0 open vision-language model from AI2. Strong on document+chart understanding.

by Allen Institute

Qwen2.5-VL 7B

Apache 2.0 7B Самоорганизующиеся

Advanced vision-language model that understands images, charts, and documents. Ask questions about any image.

by Alibaba Cloud

Explore more vision models

380+ external options via API

OCR & Document Understanding 4

Florence-2 Large

MIT 770M Самоорганизующиеся

Florence-2 — MIT. Unified vision model doing OCR, object detection, captioning, and grounding in …

GOT-OCR 2.0

Apache 2.0 580M Самоорганизующиеся

Unified OCR model that handles text, tables, charts, and mathematical equations. Far superior to …

by Alibaba Cloud

PaddleOCR

Apache 2.0 multi Самоорганизующиеся

PaddleOCR — Apache 2.0. Industrial-grade OCR supporting 80+ languages. Layout analysis, table detection, handwriting.

by PaddlePaddle

TrOCR Base

MIT 334M Самоорганизующиеся

TrOCR — MIT. Transformer-based OCR, especially strong on handwriting.

Explore more OCR models

380+ external options via API

Image Enhancement & Upscaling 5

BiRefNet v2

MIT 220M Самоорганизующиеся

BiRefNet v2 — MIT. Bilateral reference dichotomous segmentation. Sharper than U-2-Net / RMBG on …

GFPGAN

Apache 2.0 333M Самоорганизующиеся

GFPGAN — Apache 2.0. Face restoration + enhancement model. Fixes blurry/damaged portraits.

Marigold Depth

OpenRAIL++-M 865M Самоорганизующиеся

Marigold — OpenRAIL++-M. CVPR 2024 best paper monocular depth estimator, fine-tuned from Stable Diffusion. …

Real-ESRGAN

BSD Самоорганизующиеся

State-of-the-art image upscaling and enhancement.

Segment Anything 2

Apache 2.0 224M Самоорганизующиеся

SAM 2 — Apache 2.0. Meta's universal segmentation model, with video support.

Explore more models

380+ external options via API

Audio Source Separation 2

Demucs

MIT Самоорганизующиеся

Separate vocals, drums, bass, and other stems from any song.

Spleeter

MIT 20M Самоорганизующиеся

Spleeter — MIT. Stem separation (vocals, drums, bass, piano, other).

Explore more models

380+ external options via API

3D Generation 3

InstantMesh

Apache 2.0 1B Самоорганизующиеся

InstantMesh — Apache 2.0. Single-image → 3D mesh in seconds.

TRELLIS

MIT 2B Самоорганизующиеся

TRELLIS — MIT. Unified 3D generation from text or image. Industry-leading mesh quality.

TripoSR

MIT Самоорганизующиеся

Generate 3D meshes from single images in under a second. Outputs GLB or OBJ …

by VAST AI / Stability AI

Explore more models

380+ external options via API

Text Embeddings 3

BGE-M3

MIT Самоорганизующиеся

State-of-the-art multilingual text embedding model. Powers semantic search and similarity matching.

Nomic Embed v2

Apache 2.0 305M Самоорганизующиеся

Nomic Embed v2 — Apache 2.0. Matryoshka-embeddings, truncatable 768→256→128 dims.

mxbai-embed-large-v1

Apache 2.0 335M Самоорганизующиеся

mxbai-embed-large-v1 — Apache 2.0. Strong on MTEB English benchmark.

by mixedbread.ai

Explore more models

380+ external options via API

All models include a free daily token pool

Каждая модель на Free.ai - это открытый источник (MIT / Apache 2.0 / BSD). Для попробования не требуется регистрации. Выходы для коммерческого использования.

Try Free AI Tools Просмотр всех 380+ моделей

Расценить эту страницу