NEW PLUGIN EMOCHATLM
Quote from emo on May 5, 2024, 10:57 pmLAST PLUGIN VERSION:1.11
added getmodels function
added run cli features in example (read about)Now you can make API requests to your local server running LM studio from VisualNeo WIN.
LM studio :
Run LLMs on your laptop, entirely offline
Download any compatible model files from HuggingFace repositories
With the new LM studio CLI you can now:
load/unload models, start/stop the API server
LAST PLUGIN VERSION:1.11
added getmodels function
added run cli features in example (read about)
Now you can make API requests to your local server running LM studio from VisualNeo WIN.
LM studio :
Run LLMs on your laptop, entirely offline
Download any compatible model files from HuggingFace repositories
With the new LM studio CLI you can now:
load/unload models, start/stop the API server
- You need to login to have access to uploads.
Quote from luiz on June 4, 2024, 2:03 amGood work Emmanuel. Works very well. Thanks.
Some informations about the Pub interface:
- Max Tokens:
- Definition:
- Max Tokens determines the maximum size for the text generated by the model.
- By setting a limit, we can prevent the model from producing overly long texts, which could result in context loss or irrelevant responses.
- Usage:
- Max Tokens is especially useful when you want to limit the size of the generated outputs.
- For example, when creating automatic summaries or short messages, setting a low value for Max Tokens ensures concise and relevant text.
- Value -1:
- When you set Max Tokens to -1, it means there is no specific limit.
- The model will generate the complete text regardless of its length.
- Use this setting carefully, as it may result in very long and possibly irrelevant outputs.
- Temperature:
- Definition:
- Temperature is a crucial parameter when working with Large Language Models (LLMs).
- It influences the randomness and creativity of the text generated by the model (0 to 1).
- Higher Temperature values lead to more diverse and creative output.
- Conversely, lower Temperature values result in more conservative and deterministic responses.
- Value 0.2:
- A Temperature of 0.2 yields more coherent and less random outputs.
- It's a good choice when you want direct and predictable answers.
Good work Emmanuel. Works very well. Thanks.
Some informations about the Pub interface:
- Max Tokens:
- Definition:
- Max Tokens determines the maximum size for the text generated by the model.
- By setting a limit, we can prevent the model from producing overly long texts, which could result in context loss or irrelevant responses.
- Usage:
- Max Tokens is especially useful when you want to limit the size of the generated outputs.
- For example, when creating automatic summaries or short messages, setting a low value for Max Tokens ensures concise and relevant text.
- Value -1:
- When you set Max Tokens to -1, it means there is no specific limit.
- The model will generate the complete text regardless of its length.
- Use this setting carefully, as it may result in very long and possibly irrelevant outputs.
- Definition:
- Temperature:
- Definition:
- Temperature is a crucial parameter when working with Large Language Models (LLMs).
- It influences the randomness and creativity of the text generated by the model (0 to 1).
- Higher Temperature values lead to more diverse and creative output.
- Conversely, lower Temperature values result in more conservative and deterministic responses.
- Value 0.2:
- A Temperature of 0.2 yields more coherent and less random outputs.
- It's a good choice when you want direct and predictable answers.
- Definition:
Quote from emo on June 3, 2025, 10:12 amInglés: You can use the same plugin to connect to Ollama on localhost:11434, and you can even load the models since the API is fully compatible. https://ollama.com/download/windows
Español: Puedes utilizar el mismo plugin para conectarte a Ollama en localhost:11434, incluso puedes cargar los modelos ya que la API es totalmente compatible. https://ollama.com/download/windows
Inglés: You can use the same plugin to connect to Ollama on localhost:11434, and you can even load the models since the API is fully compatible. https://ollama.com/download/windows
Español: Puedes utilizar el mismo plugin para conectarte a Ollama en localhost:11434, incluso puedes cargar los modelos ya que la API es totalmente compatible. https://ollama.com/download/windows