Generative AI (GenAI): Difference between revisions

From Tygron Support wiki
Jump to navigation Jump to search
← Older edit
Maxim@tygron.com (talk | contribs)
Tygron, Tygron_admin, Bureaucrats, Interface administrators, Administrators
6,366 edits
No edit summary
No edit summary
 
(19 intermediate revisions by 2 users not shown)
Line 1: Line 1:
Large Language Models (LLM){{ref|LLM}} are a type of [[Neural Network]] in the {{software}} used by the [[Chat Channel#AI Agents|AI Agent]] to generate [[Chat Message]]s in [[Chat Channel]]s.
Generative AI [[Neural Network]] models are used in multiple ways in the {{software}} by [[AI Agent]]s.


LLMs are trained on vast amounts of text to understand and generate human-like language for tasks like summarizing, translating, answering questions and interacting with your [[Project]].
===Large Language Models===
List of open weight Large Language Models (LLM){{ref|LLM}} models available to Chat Agents.
* '''Gemma4 (Speed)''': Gemma4 model{{ref|Gemma4}} optimized for fast reactions.
* '''Qwen3.8 (Quality)''': Qwen3.8 model{{ref|Qwen3.8}} optimized for quality needed for coding, creating [[TQL]] queries and calling session endpoints.
 
===Vision Models===
These models help the LLM's to identify features in images from the interface or e.g. Overlay exports.
 
===Diffusion Models===
These models can create new or edit existing images that can be used in Design Agents.
* '''Qwen-Image 2511''': Qwen Image model{{ref|QwenImage}}.
 
===Embeddings===
Embedding model used for AI assisted data search.
* '''Embeddinggemma''': Embedding Gemma{{ref|EmbeddingGemma}} Fast embedding model.


{{article end
{{article end
|notes=
|notes=
* The {{software}} supports multiple [[Chat Channel#LLM|open-source models]], for example Gemma4{{ref|Gemma4}} and Mistral{{ref|Mistral}}.
* The LLM also has access to several [[Chat Channel#MCP Tools|Model Context Protocol (MCP) tools]] {{ref|MCP}} in order to interact with this WIKI and your [[Project]].
* All models run Tygron [[Cloud Computing]] servers, with no links to external hosting and follow strict [[AI Rules]].
* All models run Tygron [[Cloud Computing]] servers, with no links to external hosting and follow strict [[AI Rules]].
|seealso=*[[Chat Channel]]
|seealso=*[[Chat Channel]]
*[[AI Agent]]
|references=
|references=
<references>
<references>
Line 27: Line 40:
   |lastvisit=2026-06-29
   |lastvisit=2026-06-29
}}
}}
{{ref|Mistral
{{ref|EmbeddingGemma
   |name=Mistral
   |name=Embedding Gemma
   |author=
   |author=
   |page=
   |page=
   |source=
   |source=
   |link=https://mistral.ai/news/vibe-remote-agents-mistral-medium-3-5/
   |link=https://deepmind.google/models/gemma/embeddinggemma/
   |lastvisit=2026-06-29
  |lastvisit=2026-10-07
}}
{{ref|Qwen3.8
  |name=Qwen3.8
  |author=
  |page=
  |source=
  |link=https://qwen.ai/
   |lastvisit=2026-09-03
}}
}}
{{ref|MCP
{{ref|QwenImage
   |name=MCP
   |name=Qwen Image 2511
   |author=
   |author=
   |page=
   |page=
   |source=
   |source=
   |link=https://modelcontextprotocol.io/docs/getting-started/intro
   |link=https://github.com/QwenLM/Qwen-Image
   |lastvisit=2026-06-29
   |lastvisit=2026-10-03
}}
}}
</references>
</references>
|api=* [[Api session event editor chatchannel set neural network]]
|howtos=*[[How to change the LLM of a Chat Channel]]
* [[Api session event editor chatchannel set tools]]
|api=* [[Api session event editor chat set neural network]]
}}
}}
{{Editor community nav}}
{{Editor community nav}}

Latest revision as of 14:45, 7 October 2026

Generative AI Neural Network models are used in multiple ways in the Tygron Platform by AI Agents.

Large Language Models

List of open weight Large Language Models (LLM)[1] models available to Chat Agents.

  • Gemma4 (Speed): Gemma4 model[2] optimized for fast reactions.
  • Qwen3.8 (Quality): Qwen3.8 model[3] optimized for quality needed for coding, creating TQL queries and calling session endpoints.

Vision Models

These models help the LLM's to identify features in images from the interface or e.g. Overlay exports.

Diffusion Models

These models can create new or edit existing images that can be used in Design Agents.

  • Qwen-Image 2511: Qwen Image model[4].

Embeddings

Embedding model used for AI assisted data search.

  • Embeddinggemma: Embedding Gemma[5] Fast embedding model.

Notes

How-to's

See also

API Endpoints

References

  1. ↑ LLM ∙ Found at: https://en.wikipedia.org/wiki/Large_language_model ∙ (last visited: 2026-06-29)
  2. ↑ Gemma4 ∙ Found at: https://deepmind.google/models/gemma/gemma-4/ ∙ (last visited: 2026-06-29)
  3. ↑ Qwen3.8 ∙ Found at: https://qwen.ai/ ∙ (last visited: 2026-09-03)
  4. ↑ Qwen Image 2511 ∙ Found at: https://github.com/QwenLM/Qwen-Image ∙ (last visited: 2026-10-03)
  5. ↑ Embedding Gemma ∙ Found at: https://deepmind.google/models/gemma/embeddinggemma/ ∙ (last visited: 2026-10-07)
Editor Community
Community
Public Share • Blog • Wiki • Tickets • Chats • Sign in ArcGIS
GenAI
Chat Channel • Chat Message • AI Agent • Skill • Agent Tool • AI Rules • Agent Attributes
Editor
Current Situation • Future Design • Multi Scenario • Tools • Community


Retrieved from "https://support.tygron.com/w/index.php?title=Generative_AI_(GenAI)&oldid=94497"