DeepSeek-R1-webGPU single HTML/Javascript page with IndexedDB Document Storage

Open the console (Shift-Ctrl-I) for more info. This single HTML/Javascript Browser LLM is too big for your cell phone. If you don't want to completely download huggingface.co/onnx-community/DeepSeek-R1-Distill-Qwen-1.5B-ONNX then you should probably close this page.

It will load from cache if downloaded once. Uses the Web-GPU TransformersJS DeepseekR1 model or other models:

Data warning ~1.4 GB saved to cache

Click for hyperparameters which may or may not work

Max tokens in the answer:
Top_p: closer to zero more focused, closer to 1 more variety
Temperature: Close to 0 more predictable, closer to 1 more diverse

Top_k:
Min_length:
Repetition_penalty:
Length_penalty:

Do_sample: when selected, picks and considers more token options making it varied and creative, not selected keeps safest token
Chain_of_thought: AI explains more about its token creation process.
Early_stopping:

Loading progress: 0%

Rendered Output:

...

Local Document Storage (IndexedDB)

Store text documents locally in your browser. These documents can be copied into the prompt to provide context to the LLM. This is the first step towards a RAG system, but note that automated semantic retrieval (which requires an embedding model and vector search) is not included here.

Stored Documents:

Loading documents...

Use at your own risk, by Jeremy Ellis LinkedIn
Github index at hpssjellis.github.io/my-examples-of-ai-agents/public/index.html
My transformersjs github where the work was done [https://github.com/hpssjellis/my-examples-of-transformersJS](https://github.com/hpssjellis/my-examples-of-transformersJS) My Github Profile