InferVerse - Universal AI Inference Engine

🤖

Welcome to InferVerse

Select a model and start a conversation. All inference runs locally in your browser.

Model Management

Download and manage AI models for local inference

AI Tasks

Run specialized AI tasks with Transformers.js

Inference History

View past inference sessions

Generated Outputs

Images, audio, and other generated content

Settings

Configure InferVerse

Inference Settings

Temperature

0.70

Top-K

40

Top-P

0.90

Min-P

0.05

Repetition Penalty

1.10

Presence Penalty

0.00

Frequency Penalty

0.00

Max Tokens

2048

Context Length

4096

Batch Size

512

Seed

-1 for random

WASM Threads

4

System Settings

Language

Default System Prompt

Enable WebGPU

Enable Multi-threading

Enable Streaming

Storage Management

Models:0 MB

Cache:0 MB

History:0 MB

Outputs:0 MB

System Information

WebGPU:Checking...

WebAssembly:Checking...

SharedArrayBuffer:Checking...

User Agent:-