-
Notifications
You must be signed in to change notification settings - Fork 0
Models The best
Mike edited this page May 28, 2026
·
1 revision
Эта страница использует catalog heuristics: fast = минимальные vramMB/diskMB; quality = qualityRank если есть, иначе более сильная модель внутри категории.
| Категория | Лучшее на скорость/размер | Требования | Лучшее на качество/размер | Требования |
|---|---|---|---|---|
| LLM / чат и генерация текста | Qwen2.5 0.5B Instruct |
tiny, 696 MB VRAM/RAM, 220 MB disk | Meta Llama 3.1 405B Instruct |
large, 88293 MB VRAM/RAM, 48885 MB disk |
| Embeddings / векторные представления | all-MiniLM-L6-v2 |
tiny, 250 MB VRAM/RAM, 90 MB disk | Qwen3 Embedding 4B ONNX ST |
medium, 20088 MB VRAM/RAM, 15354 MB disk |
| Reranker / переранжирование | reranker bert tiny gooaq bce |
tiny, 170 MB VRAM/RAM, 33 MB disk | Qwen3 Reranker 0.6B |
tiny, 10559 MB VRAM/RAM, 8024 MB disk |
| Translator / перевод | OPUS MT EN-RU |
tiny, 650 MB VRAM/RAM, 310 MB disk | NLLB 200 distilled 600M |
medium, 2300 MB VRAM/RAM, 1200 MB disk |
| TTS / синтез речи | mms tts deu |
tiny, 256 MB VRAM/RAM, 180 MB disk | Kokoro 82M |
tiny, 450 MB VRAM/RAM, 180 MB disk |
| VLM / image-to-text | SmolVLM 256M Instruct |
tiny, 291 MB VRAM/RAM, 243 MB disk | Qwen2 VL 2B Instruct |
medium, 2280 MB VRAM/RAM, 1900 MB disk |
| ASR / speech-to-text | Whisper tiny multilingual |
tiny, 250 MB VRAM/RAM, 80 MB disk | Whisper large v3 turbo |
large, 3600 MB VRAM/RAM, 1600 MB disk |
| OCR / распознавание текста | trocr base stage1 |
tiny, 312 MB VRAM/RAM, 260 MB disk | trocr base stage1 |
tiny, 312 MB VRAM/RAM, 260 MB disk |
| Image classification / классификация изображений | MobileNet V2 |
tiny, 160 MB VRAM/RAM, 25 MB disk | ViT base 224 |
small, 700 MB VRAM/RAM, 330 MB disk |
| Object detection / детекция объектов | detr resnet 50 panoptic |
tiny, 256 MB VRAM/RAM, 180 MB disk | DETR ResNet 50 |
medium, 900 MB VRAM/RAM, 170 MB disk |
| Image segmentation / сегментация и background removal | MODNet portrait matting |
tiny, 180 MB VRAM/RAM, 30 MB disk | BEN2 background removal |
medium, 1300 MB VRAM/RAM, 680 MB disk |
| Depth estimation / карта глубины | depth anything base hf |
tiny, 264 MB VRAM/RAM, 220 MB disk | DPT hybrid MiDaS |
medium, 950 MB VRAM/RAM, 480 MB disk |
| Document layout / разметка документа | yolov10b doclaynet |
tiny, 256 MB VRAM/RAM, 180 MB disk | yolov10b doclaynet |
tiny, 256 MB VRAM/RAM, 180 MB disk |
| Table detection / таблицы | table transformer detection |
tiny, 256 MB VRAM/RAM, 180 MB disk | table transformer structure recognition v1.1 all |
tiny, 256 MB VRAM/RAM, 180 MB disk |
| Document QA / вопрос-ответ по документу | donut base finetuned cord v2 |
small, 504 MB VRAM/RAM, 420 MB disk | donut base finetuned cord v2 |
small, 504 MB VRAM/RAM, 420 MB disk |
| Language ID / определение языка | mms lid 126 |
tiny, 256 MB VRAM/RAM, 120 MB disk | mms lid 4017 |
tiny, 256 MB VRAM/RAM, 120 MB disk |
| Audio classification / классификация аудио | ast finetuned audioset 10 10 0.4593 |
tiny, 256 MB VRAM/RAM, 180 MB disk | wav2vec2 large xlsr 53 gender recognition librispeech |
tiny, 256 MB VRAM/RAM, 180 MB disk |
| Text classification / классификация текста | bert base multilingual uncased sentiment |
tiny, 256 MB VRAM/RAM, 120 MB disk | toxic bert |
tiny, 256 MB VRAM/RAM, 120 MB disk |
| NER / извлечение сущностей | bert base NER |
tiny, 256 MB VRAM/RAM, 120 MB disk | bert base multilingual cased ner hrl |
tiny, 256 MB VRAM/RAM, 120 MB disk |
| Zero-shot text / классификация текста по меткам | bart large mnli |
small, 396 MB VRAM/RAM, 330 MB disk | mobilebert uncased mnli |
small, 396 MB VRAM/RAM, 330 MB disk |
| Summarization / суммаризация | Bart Large CNN |
small, 504 MB VRAM/RAM, 420 MB disk | text summarization |
small, 504 MB VRAM/RAM, 420 MB disk |
| Text2Text / text-to-text задачи | LaMini T5 61M |
tiny, 256 MB VRAM/RAM, 57 MB disk | LaMini Flan T5 783M |
small, 891 MB VRAM/RAM, 743 MB disk |
| Code understanding / кодовые embeddings | CodeBERTa language id |
tiny, 256 MB VRAM/RAM, 160 MB disk | codebert javascript |
tiny, 256 MB VRAM/RAM, 160 MB disk |
| Категория | Лучшее на скорость/размер | Требования | Лучшее на качество/размер | Требования |
|---|---|---|---|---|
| LLM / чат и генерация текста | SmolLM2 360M Instruct |
tiny, 700 MB VRAM/RAM, 280 MB disk | gpt oss 20b |
large, 34200 MB VRAM/RAM, 19000 MB disk |
| Embeddings / векторные представления | Snowflake Arctic Embed L v2 |
large, 2400 MB VRAM/RAM, 1200 MB disk | Mixedbread embed large |
large, 2600 MB VRAM/RAM, 1350 MB disk |
| Reranker / переранжирование | BGE Reranker v2 M3 |
medium, 1600 MB VRAM/RAM, 900 MB disk | BGE Reranker v2 M3 |
medium, 1600 MB VRAM/RAM, 900 MB disk |
| Translator / перевод | M2M100 418M |
medium, 1900 MB VRAM/RAM, 900 MB disk | NLLB 200 distilled 600M |
medium, 2300 MB VRAM/RAM, 1200 MB disk |
| VLM / image-to-text | Qwen2 VL 2B Instruct |
medium, 2280 MB VRAM/RAM, 1900 MB disk | Qwen2 VL 2B Instruct |
medium, 2280 MB VRAM/RAM, 1900 MB disk |
| ASR / speech-to-text | Whisper large v3 turbo |
large, 3600 MB VRAM/RAM, 1600 MB disk | Whisper large v3 turbo |
large, 3600 MB VRAM/RAM, 1600 MB disk |
| Категория | Лучшее на скорость/размер | Требования | Лучшее на качество/размер | Требования |
|---|---|---|---|---|
| LLM / чат и генерация текста | SmolLM 135M |
tiny, 256 MB VRAM/RAM, 128 MB disk | Llama 3.2 1B Instruct |
small, 1710 MB VRAM/RAM, 950 MB disk |
| Embeddings / векторные представления | all-MiniLM-L6-v2 |
tiny, 250 MB VRAM/RAM, 90 MB disk | Qwen3 Embedding 0.6B |
small, 684 MB VRAM/RAM, 570 MB disk |
| Reranker / переранжирование | bge reranker large |
tiny, 360 MB VRAM/RAM, 300 MB disk | Qwen3 Reranker 0.6B |
small, 684 MB VRAM/RAM, 570 MB disk |
| Translator / перевод | OPUS MT EN-RU |
tiny, 650 MB VRAM/RAM, 310 MB disk | opus mt de en |
small, 384 MB VRAM/RAM, 320 MB disk |
| TTS / синтез речи | mms tts deu |
tiny, 256 MB VRAM/RAM, 180 MB disk | Kokoro 82M |
tiny, 450 MB VRAM/RAM, 180 MB disk |
| VLM / image-to-text | SmolVLM 256M Instruct |
tiny, 291 MB VRAM/RAM, 243 MB disk | FastVLM 0.5B |
small, 570 MB VRAM/RAM, 475 MB disk |
| ASR / speech-to-text | Whisper tiny multilingual |
tiny, 250 MB VRAM/RAM, 80 MB disk | Whisper base multilingual |
small, 550 MB VRAM/RAM, 145 MB disk |
| OCR / распознавание текста | TexTeller |
tiny, 312 MB VRAM/RAM, 260 MB disk | TexTeller |
tiny, 312 MB VRAM/RAM, 260 MB disk |
| Image classification / классификация изображений | MobileNet V2 |
tiny, 160 MB VRAM/RAM, 25 MB disk | ViT base 224 |
small, 700 MB VRAM/RAM, 330 MB disk |
| Object detection / детекция объектов | detr resnet 50 panoptic |
tiny, 256 MB VRAM/RAM, 180 MB disk | DETR ResNet 50 |
medium, 900 MB VRAM/RAM, 170 MB disk |
| Image segmentation / сегментация и background removal | MODNet portrait matting |
tiny, 180 MB VRAM/RAM, 30 MB disk | BEN2 background removal |
medium, 1300 MB VRAM/RAM, 680 MB disk |
| Depth estimation / карта глубины | depth anything base hf |
tiny, 264 MB VRAM/RAM, 220 MB disk | DPT hybrid MiDaS |
medium, 950 MB VRAM/RAM, 480 MB disk |
| Document layout / разметка документа | yolov10b doclaynet |
tiny, 256 MB VRAM/RAM, 180 MB disk | yolov10b doclaynet |
tiny, 256 MB VRAM/RAM, 180 MB disk |
| Table detection / таблицы | table transformer detection |
tiny, 256 MB VRAM/RAM, 180 MB disk | table transformer detection |
tiny, 256 MB VRAM/RAM, 180 MB disk |
| Document QA / вопрос-ответ по документу | donut base finetuned cord v2 |
small, 504 MB VRAM/RAM, 420 MB disk | donut base finetuned cord v2 |
small, 504 MB VRAM/RAM, 420 MB disk |
| Zero-shot image / классификация изображения по меткам | CLIP ViT base patch32 |
small, 750 MB VRAM/RAM, 340 MB disk | SigLIP base 224 |
medium, 1100 MB VRAM/RAM, 520 MB disk |
| Language ID / определение языка | mms lid 126 |
tiny, 256 MB VRAM/RAM, 120 MB disk | mms lid 126 |
tiny, 256 MB VRAM/RAM, 120 MB disk |
| Audio classification / классификация аудио | ast finetuned audioset 10 10 0.4593 |
tiny, 256 MB VRAM/RAM, 180 MB disk | ast finetuned audioset 10 10 0.4593 |
tiny, 256 MB VRAM/RAM, 180 MB disk |
| Text classification / классификация текста | bert base multilingual uncased sentiment |
tiny, 256 MB VRAM/RAM, 120 MB disk | bert base multilingual uncased sentiment |
tiny, 256 MB VRAM/RAM, 120 MB disk |
| NER / извлечение сущностей | bert base NER |
tiny, 256 MB VRAM/RAM, 120 MB disk | bert base NER |
tiny, 256 MB VRAM/RAM, 120 MB disk |
| Zero-shot text / классификация текста по меткам | bart large mnli |
small, 396 MB VRAM/RAM, 330 MB disk | bart large mnli |
small, 396 MB VRAM/RAM, 330 MB disk |
| Summarization / суммаризация | Bart Large CNN |
small, 504 MB VRAM/RAM, 420 MB disk | Bart Large CNN |
small, 504 MB VRAM/RAM, 420 MB disk |
| Text2Text / text-to-text задачи | LaMini T5 61M |
tiny, 256 MB VRAM/RAM, 57 MB disk | LaMini Flan T5 783M |
small, 891 MB VRAM/RAM, 743 MB disk |
| Code understanding / кодовые embeddings | CodeBERTa language id |
tiny, 256 MB VRAM/RAM, 160 MB disk | CodeBERTa language id |
tiny, 256 MB VRAM/RAM, 160 MB disk |
- xlocllm
- Quickstart
- About
- Functions Python
- Functions TypeScript
- Use cases
- Examples Python
- Examples TypeScript
- Shared GPU mode
-
Models catalog
- Models The best
- Models Full model list
- Models Use your model
- For native mode
- Models Native LLM tiny small
- Models Native LLM medium
- Models Native LLM large
- Models Native embedding
- Models Native reranker
- Models Native translator
- Models Native tts
- Models Native vlm
- Models Native asr
- Models Native ocr
- Models Native image-classification
- Models Native object-detection
- Models Native image-segmentation
- Models Native depth-estimation
- Models Native document-layout
- Models Native table-detection
- Models Native document-qa
- Models Native language-id
- Models Native audio-classification
- Models Native text-classification
- Models Native ner
- Models Native zero-shot-text
- Models Native summarization
- Models Native text2text
- Models Native code
- For webgpu mode
- For web mode
- Models Web LLM
- Models Web embedding
- Models Web reranker
- Models Web translator
- Models Web tts
- Models Web vlm
- Models Web asr
- Models Web ocr
- Models Web image-classification
- Models Web object-detection
- Models Web image-segmentation
- Models Web depth-estimation
- Models Web document-layout
- Models Web table-detection
- Models Web document-qa
- Models Web zero-shot-image
- Models Web language-id
- Models Web audio-classification
- Models Web text-classification
- Models Web ner
- Models Web zero-shot-text
- Models Web summarization
- Models Web text2text
- Models Web code
- Dev