🏆 Transformers.js Benchmark Leaderboard

Compare benchmark results for different models, platforms, and configurations.

💡 Tip: Use the recommended models section below to find popular models that are fast to load and quick to run - perfect for getting started!

⭐ Recommended WebGPU Models for Beginners

These models are selected for being:

WebGPU compatible - Work in modern browsers with GPU acceleration
Beginner-friendly - Popular, fast to load, and quick to run
Sorted by task type, showing top 3-5 models per task

Top WebGPU-Compatible Models by Task

Top WebGPU-Compatible Models by Task

bd64ed68-37bc-4f7c-a0bf-6b02a17fbc50	🌐 web	onnx-community/dinov3-vits16-pretrain-lvd1689m-ONNX-MHA-scores	automatic-speech-recognition	🔥 warm	3	1	⚡ webgpu	🔵 chromium	fp32	No	✅ completed	2025-10-21 09:11:50	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	🐌 10079.9ms	🐌 14340.1ms	🐌 1314.3ms	🐌 1320.1ms	🐌 1251.4ms	🐌 1272.2ms	💪 32.0 cores	⚠️ 8.0GB	⚠️ 162.2s	📊 10.0k	💚 19	⭐⭐⭐ 90

📝 Markdown Output for llms.txt

Copy the markdown below to embed in your llms.txt or documentation:

Markdown for llms.txt

# Recommended Transformers.js Models for First-Time Trials

This guide provides curated model recommendations for each task type, selected for their:
- **Popularity**: Widely used with strong community support
- **Performance**: Fast loading and inference times
- **WebGPU Compatibility**: GPU-accelerated in modern browsers

**Important:** These recommendations are designed for initial experimentation and learning. Many other models are available for each task. **You should evaluate and choose the best model for your specific use case, performance requirements, and constraints.**

## About the Model Recommendations

The models below are selected for their popularity and ease of use, making them ideal for initial experimentation. **This list does not cover all available models** - you should evaluate and select the best model for your specific use case and requirements.

## Audio-Classification

**Usage Example:** Perform audio classification

```javascript
const classifier = await pipeline('audio-classification', 'onnx-community/Musical-genres-Classification-Hubert-V1-ONNX');
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/jfk.wav';
const output = await classifier(url);
```

### Recommended Models for First-Time Trials

#### onnx-community/Musical-genres-Classification-Hubert-V1-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 264.4ms | Inference: 136.5ms | Downloads: 6 | Likes: 2

#### Xenova/ast-finetuned-audioset-10-10-0.4593

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 2827.1ms | Inference: 544.1ms | Downloads: 45

#### onnx-community/Speech-Emotion-Classification-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3070.4ms | Inference: 574.9ms | Downloads: 7 | Likes: 2

#### onnx-community/ast-finetuned-audioset-10-10-0.4593-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3395.4ms | Inference: 565.2ms | Downloads: 14 | Likes: 1

#### Xenova/wav2vec2-base-superb-ks

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3211.4ms | Inference: 584.7ms | Downloads: 12 | Likes: 1

---

## Automatic-Speech-Recognition

**Usage Example:** Transcribe audio from a URL

```javascript
const transcriber = await pipeline('automatic-speech-recognition', 'Xenova/whisper-tiny.en');
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/jfk.wav';
const output = await transcriber(url);
```

### Recommended Models for First-Time Trials

#### Xenova/whisper-tiny.en

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1045.6ms | Inference: 1314.3ms | Downloads: 43.6k | Likes: 19

#### onnx-community/moonshine-base-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1906.3ms | Inference: 889.1ms | Downloads: 10.7k | Likes: 30

#### Xenova/whisper-tiny

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1037.2ms | Inference: 1258.4ms | Downloads: 23.9k | Likes: 10

#### onnx-community/whisper-base

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1885.8ms | Inference: 878.9ms | Downloads: 15.4k | Likes: 22

#### Xenova/whisper-base.en

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1591.3ms | Inference: 1941.5ms | Downloads: 33.2k | Likes: 1

---

## Depth-Estimation

**Usage Example:** Depth estimation

```javascript
const depth_estimator = await pipeline('depth-estimation', 'Xenova/depth-anything-small-hf');
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/cats.jpg';
const out = await depth_estimator(url);
```

### Recommended Models for First-Time Trials

#### Xenova/depth-anything-small-hf

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 414.3ms | Inference: 419.3ms | Downloads: 3.0k | Likes: 11

#### onnx-community/depth-anything-v2-small

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 450.0ms | Inference: 357.1ms | Downloads: 1.5k | Likes: 22

#### onnx-community/metric3d-vit-small

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 786.2ms | Inference: 515.7ms | Downloads: 17 | Likes: 2

#### Xenova/glpn-nyu

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1454.5ms | Inference: 1076.9ms | Downloads: 6

#### Xenova/glpn-kitti

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1451.4ms | Inference: 1083.3ms | Downloads: 6

---

## Feature-Extraction

**Usage Example:** Run feature extraction

```javascript
const extractor = await pipeline('feature-extraction', 'Xenova/all-MiniLM-L6-v2');
const output = await extractor('This is a simple test.');
```

### Recommended Models for First-Time Trials

#### Xenova/all-MiniLM-L6-v2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 373.5ms | Inference: 143.9ms | Downloads: 432.6k | Likes: 87

#### Xenova/bge-base-en-v1.5

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3620.8ms | Inference: 816.4ms | Downloads: 1.1M | Likes: 8

#### Xenova/gte-small

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 588.7ms | Inference: 228.4ms | Downloads: 10.0k | Likes: 22

#### onnx-community/bert_uncased_L-2_H-128_A-2-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 110.8ms | Inference: 49.2ms | Downloads: 8

#### Xenova/bge-small-en-v1.5

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 555.1ms | Inference: 247.2ms | Downloads: 38.1k | Likes: 11

---

## Fill-Mask

**Usage Example:** Perform masked language modelling (a.k.a. "fill-mask")

```javascript
const unmasker = await pipeline('fill-mask', 'Xenova/distilbert-base-uncased');
const output = await unmasker('The goal of life is [MASK].');
```

### Recommended Models for First-Time Trials

#### Xenova/distilbert-base-uncased

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 769.8ms | Inference: 302.4ms | Downloads: 362 | Likes: 1

#### Xenova/bert-base-uncased

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 4432.7ms | Inference: 919.4ms | Downloads: 601 | Likes: 1

#### Xenova/albert-base-v2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 324.9ms | Inference: 68.1ms | Downloads: 23 | Likes: 1

#### Xenova/distilbert-base-cased

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 742.7ms | Inference: 301.1ms | Downloads: 20 | Likes: 1

#### Xenova/albert-large-v2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 499.6ms | Inference: 78.8ms | Downloads: 15

---

## Image-Classification

**Usage Example:** Classify an image

```javascript
const classifier = await pipeline('image-classification', 'Xenova/vit-base-patch16-224');
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/tiger.jpg';
const output = await classifier(url);
```

### Recommended Models for First-Time Trials

#### Xenova/vit-base-patch16-224

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 2894.1ms | Inference: 476.3ms | Downloads: 4.4k | Likes: 2

#### Xenova/facial_emotions_image_detection

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 2922.3ms | Inference: 479.3ms | Downloads: 243 | Likes: 6

#### Xenova/resnet-50

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 304.9ms | Inference: 160.5ms | Downloads: 774

#### Xenova/resnet-18

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 152.1ms | Inference: 90.8ms | Downloads: 13 | Likes: 2

#### onnx-community/mobilenet_v2_1.0_224

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 90.7ms | Inference: 99.7ms | Downloads: 54 | Likes: 1

---

## Image-Feature-Extraction

**Usage Example:** Perform image feature extraction

```javascript
const image_feature_extractor = await pipeline('image-feature-extraction', 'onnx-community/dinov3-vits16-pretrain-lvd1689m-ONNX');
const url = 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/cats.png';
const features = await image_feature_extractor(url);
```

### Recommended Models for First-Time Trials

#### onnx-community/dinov3-vits16-pretrain-lvd1689m-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 311.4ms | Inference: 352.4ms | Downloads: 1.6k | Likes: 11

#### onnx-community/dinov3-vits16-pretrain-lvd1689m-ONNX-MHA-scores

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 342.8ms | Inference: 259.6ms | Downloads: 516 | Likes: 2

#### Xenova/dino-vits16

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 321.7ms | Inference: 247.3ms | Downloads: 640

#### onnx-community/dinov2-with-registers-small

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 312.4ms | Inference: 270.6ms | Downloads: 4 | Likes: 1

#### onnx-community/dinov2-small

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 327.9ms | Inference: 275.2ms | Downloads: 7 | Likes: 1

---

## Image-Segmentation

**Usage Example:** Perform image segmentation

```javascript
const segmenter = await pipeline('image-segmentation', 'Xenova/segformer-b0-finetuned-ade-512-512');
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/cats.jpg';
const output = await segmenter(url);
```

### Recommended Models for First-Time Trials

#### Xenova/segformer-b0-finetuned-ade-512-512

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 205.6ms | Inference: 1612.3ms | Downloads: 2.4M | Likes: 1

#### Xenova/modnet

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 144.6ms | Inference: 286.9ms | Downloads: 7.6k | Likes: 61

#### onnx-community/modnet-webnn

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 147.5ms | Inference: 223.2ms | Downloads: 4 | Likes: 3

#### Xenova/segformer_b0_clothes

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 203.9ms | Inference: 632.4ms | Downloads: 149 | Likes: 2

#### Xenova/segformer-b0-finetuned-cityscapes-1024-1024

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 200.2ms | Inference: 519.0ms | Downloads: 3

---

## Image-To-Image

**Usage Example:**

```javascript
const processor = await pipeline('image-to-image', 'Xenova/2x_APISR_RRDB_GAN_generator-onnx');
const url = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/tiger.jpg';
const output = await processor(url);
```

### Recommended Models for First-Time Trials

#### Xenova/2x_APISR_RRDB_GAN_generator-onnx

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 117.5ms | Inference: 596.3ms | Downloads: 17 | Likes: 1

#### Xenova/4x_APISR_GRL_GAN_generator-onnx

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 2712.5ms | Inference: 3362.6ms | Downloads: 24 | Likes: 11

---

## Object-Detection

**Usage Example:** Run object-detection

```javascript
const detector = await pipeline('object-detection', 'Xenova/detr-resnet-50');
const img = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/cats.jpg';
const output = await detector(img, { threshold: 0.9 });
```

### Recommended Models for First-Time Trials

#### Xenova/detr-resnet-50

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 718.2ms | Inference: 625.4ms | Downloads: 6.4k | Likes: 16

#### Xenova/yolos-tiny

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 190.1ms | Inference: 904.3ms | Downloads: 1.7k | Likes: 6

#### onnx-community/rfdetr_base-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 470.4ms | Inference: 420.9ms | Downloads: 7 | Likes: 2

#### Xenova/yolos-small-300

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 400.0ms | Inference: 1776.1ms | Downloads: 241

#### onnx-community/rfdetr_medium-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 578.1ms | Inference: 486.1ms | Downloads: 13 | Likes: 1

---

## Question-Answering

**Usage Example:** Run question answering

```javascript
const answerer = await pipeline('question-answering', 'Xenova/distilbert-base-cased-distilled-squad');
const question = 'Who was Jim Henson?';
const context = 'Jim Henson was a nice puppet.';
const output = await answerer(question, context);
```

### Recommended Models for First-Time Trials

#### Xenova/distilbert-base-cased-distilled-squad

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 746.8ms | Inference: 164.3ms | Downloads: 6.9k | Likes: 4

#### Xenova/distilbert-base-uncased-distilled-squad

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 726.1ms | Inference: 204.5ms | Downloads: 1.0k

#### onnx-community/rubert_tiny_qa_kontur-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 411.4ms | Inference: 111.1ms | Downloads: 5

#### onnx-community/mobilebert-uncased-squad-v2-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 465.3ms | Inference: 911.1ms | Downloads: 5

#### onnx-community/bert-base-uncased-squad2-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 4776.9ms | Inference: 837.5ms | Downloads: 3 | Likes: 1

---

## Summarization

**Usage Example:** Summarization

```javascript
const generator = await pipeline('summarization', 'Xenova/distilbart-cnn-6-6');
const text = 'The tower is 324 metres (1,063 ft) tall, about the same height as an 81-storey building, ' +
  'and the tallest structure in Paris. Its base is square, measuring 125 metres (410 ft) on each side. ' +
  'During its construction, the Eiffel Tower surpassed the Washington Monument to become the tallest ' +
  'man-made structure in the world, a title it held for 41 years until the Chrysler Building in New ' +
  'York City was finished in 1930. It was the first structure to reach a height of 300 metres. Due to ' +
  'the addition of a broadcasting aerial at the top of the tower in 1957, it is now taller than the ' +
  'Chrysler Building by 5.2 metres (17 ft). Excluding transmitters, the Eiffel Tower is the second ' +
  'tallest free-standing structure in France after the Millau Viaduct.';
const output = await generator(text, {
  max_new_tokens: 100,
});
```

### Recommended Models for First-Time Trials

#### Xenova/distilbart-cnn-6-6

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 9311.4ms | Inference: 2527.9ms | Downloads: 4.1k | Likes: 8

#### Xenova/distilbart-xsum-12-1

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 8184.2ms | Inference: 1283.7ms | Downloads: 34

#### Xenova/distilbart-xsum-6-6

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 8895.0ms | Inference: 1519.4ms | Downloads: 32 | Likes: 1

#### Xenova/distilbart-xsum-9-6

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 10079.9ms | Inference: 1730.8ms | Downloads: 28

#### Xenova/distilbart-cnn-12-3

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 8878.6ms | Inference: 2479.6ms | Downloads: 57

---

## Text-Classification

**Usage Example:**

```javascript
const classifier = await pipeline('text-classification', 'Xenova/distilbert-base-uncased-finetuned-sst-2-english');
const output = await classifier('I love transformers!');
```

### Recommended Models for First-Time Trials

#### Xenova/distilbert-base-uncased-finetuned-sst-2-english

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 662.4ms | Inference: 245.3ms | Downloads: 8.8k | Likes: 11

#### Xenova/ms-marco-MiniLM-L-6-v2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 365.1ms | Inference: 146.6ms | Downloads: 6.0k | Likes: 4

#### Xenova/ms-marco-TinyBERT-L-2-v2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 115.1ms | Inference: 56.2ms | Downloads: 1.6k | Likes: 3

#### onnx-community/language_detection-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 338.1ms | Inference: 116.0ms | Downloads: 19 | Likes: 4

#### Xenova/toxic-bert

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3789.0ms | Inference: 808.8ms | Downloads: 4.0k | Likes: 6

---

## Text-Generation

**Usage Example:** Text generation

```javascript
const generator = await pipeline('text-generation', 'Xenova/distilgpt2');
const output = await generator('Once upon a time, there was', { max_new_tokens: 10 });
```

### Recommended Models for First-Time Trials

#### Xenova/distilgpt2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3207.5ms | Inference: 469.7ms | Downloads: 7.3k | Likes: 9

#### Xenova/llama2.c-stories15M

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 396.8ms | Inference: 251.4ms | Downloads: 1.6k | Likes: 7

#### Xenova/gpt2

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 5533.3ms | Inference: 1395.6ms | Downloads: 3.4k | Likes: 9

#### onnx-community/ettin-decoder-17m-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 328.2ms | Inference: 221.9ms | Downloads: 15 | Likes: 1

#### onnx-community/Lamina-extend-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 141.9ms | Inference: 190.0ms | Downloads: 14

---

## Token-Classification

**Usage Example:** Perform named entity recognition

```javascript
const classifier = await pipeline('token-classification', 'onnx-community/NeuroBERT-NER-ONNX');
const output = await classifier('My name is Sarah and I live in London');
```

### Recommended Models for First-Time Trials

#### onnx-community/NeuroBERT-NER-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 170.4ms | Inference: 61.6ms | Downloads: 14 | Likes: 2

#### onnx-community/TinyBERT-finetuned-NER-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 261.7ms | Inference: 79.9ms | Downloads: 6 | Likes: 1

#### Xenova/esm2_t6_8M_UR50D_rna_binding_site_predictor

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 279.3ms | Inference: 118.6ms | Downloads: 6 | Likes: 1

#### onnx-community/distilbert-NER-ONNX

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 682.0ms | Inference: 258.2ms | Downloads: 11 | Likes: 2

#### Xenova/bert-base-NER

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 3621.3ms | Inference: 814.7ms | Downloads: 1.8k | Likes: 1

---

## Translation

**Usage Example:** Multilingual translation

```javascript
const translator = await pipeline('translation', 'Xenova/opus-mt-en-es');
const output = await translator('Life is like a box of chocolate.', {
  src_lang: '...',
  tgt_lang: '...',
});
```

### Recommended Models for First-Time Trials

#### Xenova/opus-mt-en-es

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1956.5ms | Inference: 617.7ms | Downloads: 2.0k | Likes: 3

#### Xenova/opus-mt-en-fr

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1786.1ms | Inference: 446.7ms | Downloads: 542 | Likes: 2

#### Xenova/opus-mt-zh-en

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1964.7ms | Inference: 799.0ms | Downloads: 59 | Likes: 4

#### Xenova/opus-mt-en-zh

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1830.0ms | Inference: 481.5ms | Downloads: 493 | Likes: 2

#### Xenova/opus-mt-ar-en

**WebGPU Compatible:** ✅ Yes

**Metrics:** Load: 1756.6ms | Inference: 733.8ms | Downloads: 236 | Likes: 2

---

## About These Recommendations

### Selection Criteria

Models in this guide are selected based on:
- **Popularity**: High download counts and community engagement on HuggingFace Hub
- **Performance**: Fast loading and inference times based on benchmark results
- **Compatibility**: Verified WebGPU support for GPU-accelerated browser execution

### Understanding Benchmark Metrics

**Important:** All performance metrics (load time, inference time, etc.) are measured in a controlled benchmark environment. These metrics are useful for **comparing models against each other**, but they may not reflect the actual performance you'll experience in your specific environment. Factors that affect real-world performance include:
- Hardware specifications (CPU, GPU, memory)
- Browser type and version
- Operating system
- Network conditions (for model loading)
- Concurrent processes and system load

**We recommend** benchmarking models in your own environment with your actual use case to get accurate performance measurements.

### For Production Use

These recommendations are optimized for first-time trials and learning. For production applications, consider:
- Evaluating multiple models for your specific use case
- Testing with your actual data and performance requirements
- Reviewing the full benchmark results for comprehensive comparisons
- Exploring specialized models that may better fit your needs

Visit the full leaderboard to explore all available models and their benchmark results.

🔍 Full Benchmark Results

Model Name

Task

Platform

Device

Mode

DType

Status

All Benchmark Results

All Benchmark Results

bbd3d7b6-9714-4ba2-b78e-5caefe0a6b83	🌐 web	onnx-community/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-mammalian_lithe_barracuda-ONNX	zero-shot-image-classification	🔥 warm	3	1	⚡ webgpu	🔵 chromium	fp32	No	✅ completed	2025-10-20 14:11:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	🐌 15989.2ms	🐌 16659.4ms	🐌 27953.8ms	🐌 28033.7ms	🐌 27962.2ms	🐌 28086.3ms	💪 32.0 cores	⚠️ 8.0GB	🐌 367.6s	📊 10.0k	❤️ 152	⭐⭐⭐ 84

📊 Metrics

Benchmark Metrics:

load_ms: Model loading time in milliseconds
first_infer_ms: First inference time in milliseconds
subsequent_infer_ms: Subsequent inference time in milliseconds
p50/p90: 50th and 90th percentile values

HuggingFace Metrics:

downloads: Total downloads from HuggingFace Hub
likes: Number of likes on HuggingFace Hub

WebGPU Compatibility:

Models in the recommended section are all WebGPU compatible
WebGPU enables GPU acceleration in modern browsers

⚠️ Important Note About Performance Metrics: All metrics are measured in a controlled benchmark environment. They are useful for comparing models against each other, but may not reflect actual performance in your environment. Factors like hardware, browser, OS, and system load affect real-world performance. We recommend testing models in your own environment for accurate measurements.


bbd3d7b6-9714-4ba2-b78e-5caefe0a6b83	🌐 web	Xenova/2x_APISR_RRDB_GAN_generator-onnx	image-to-image	🔥 warm	3	1	📦 wasm	🔵 chromium	fp32	No	✅ completed	2025-10-20 14:11:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	🚀 82.4ms	🚀 85.9ms	🐌 27953.8ms	🐌 28033.7ms	🐌 27962.2ms	🐌 28086.3ms	💪 32.0 cores	⚠️ 8.0GB	🐌 367.6s	📈 17	🤍 1	· 36
51daf764-3116-4820-ac8a-a35fd2175419	🌐 web	Xenova/2x_APISR_RRDB_GAN_generator-onnx	image-to-image	🔥 warm	3	1	⚡ webgpu	🔵 chromium	fp32	No	✅ completed	2025-10-20 14:11:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	⚡ 117.5ms	⚡ 132.1ms	🐌 596.3ms	🐌 604.5ms	🐌 480.2ms	🐌 511.2ms	💪 32.0 cores	⚠️ 8.0GB	⚡ 15.0s	📈 17	🤍 1	⭐⭐ 65
12f7c420-9d7a-4b8b-9af4-649f37d986dc	🌐 web	Xenova/4x_APISR_GRL_GAN_generator-onnx	image-to-image	🔥 warm	3	1	⚡ webgpu	🔵 chromium	fp32	No	✅ completed	2025-10-20 14:11:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	⚠️ 2712.5ms	⚠️ 2717.9ms	🐌 3362.6ms	🐌 3363.8ms	🐌 3039.7ms	🐌 3070.4ms	💪 32.0 cores	⚠️ 8.0GB	✅ 59.4s	📈 24	💚 11	⭐ 49
cacadcfa-de6b-4f4a-b264-c63fa289cbf3	🌐 web	Xenova/Bio_ClinicalBERT	fill-mask	🔥 warm	3	1	📦 wasm	🔵 chromium	fp32	No	✅ completed	2025-10-17 05:51:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	⚠️ 4065.3ms	⚠️ 4155.4ms	⚠️ 97.8ms	⚠️ 99.1ms	⚠️ 95.6ms	⚠️ 97.6ms	💪 32.0 cores	⚠️ 8.0GB	✅ 20.8s	📈 9	-	⭐ 48
a7ed316d-2d69-4675-8fcd-d9888be7d155	🌐 web	Xenova/Bio_ClinicalBERT	fill-mask	🔥 warm	3	1	⚡ webgpu	🔵 chromium	fp32	No	✅ completed	2025-10-17 05:51:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	⚠️ 3909.9ms	⚠️ 4010.2ms	🐌 920.9ms	🐌 966.6ms	⚠️ 64.0ms	⚠️ 66.1ms	💪 32.0 cores	⚠️ 8.0GB	✅ 36.3s	📈 9	-	· 37
7e0866e1-857e-4471-9b34-bbc05585183c	🌐 web	Xenova/ClinicalBERT	feature-extraction	🔥 warm	3	1	📦 wasm	🔵 chromium	fp32	No	❌ failed	2025-10-20 14:11:28	Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) HeadlessChrome/140.0.7339.186 Safari/537.36	🐌 nanms	🐌 nanms	🐌 nanms	🐌 nanms	🐌 nanms	🐌 nanms	💪 32.0 cores	⚠️ 8.0GB	✅ 24.7s	📈 20	🤍 2	· nan