Z AI Character Card Wiki

The Last Letter in Personalized Artificial Intelligence

User Tools

Site Tools


backends

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
backends [2026/03/17 16:34] tysbackends [2026/03/17 16:44] (current) tys
Line 11: Line 11:
 Additionally, each of these require a separate [[models|model]] to operate. Additionally, each of these require a separate [[models|model]] to operate.
  
-=== Inference Engines ===+==== Inference Engines ====
 Each of the following perform inference without needing additional software. User friendliness is not the first priority with these utilities. Each of the following perform inference without needing additional software. User friendliness is not the first priority with these utilities.
  
 ^ Name ^ Notes ^ ^ Name ^ Notes ^
 +|  **Oft-Recommended Engines**  ||
 | [[https://github.com/ggml-org/llama.cpp|llama.cpp]] | Reference backend. Invented GGUF format. | | [[https://github.com/ggml-org/llama.cpp|llama.cpp]] | Reference backend. Invented GGUF format. |
-| [[https://github.com/LostRuins/koboldcpp|koboldcpp]] | Based on llama.cpp. Good UI. focused on RP. |+| [[https://github.com/LostRuins/koboldcpp|koboldcpp]] | Based on llama.cpp with RP focus
 +|  **Other Engines**  ||
 | [[https://github.com/ikawrakow/ik_llama.cpp|ExLlamaV3]] | Created exl3 format, focused on GPU performance. | | [[https://github.com/ikawrakow/ik_llama.cpp|ExLlamaV3]] | Created exl3 format, focused on GPU performance. |
 | [[https://github.com/ikawrakow/ik_llama.cpp|ik_llama]] | Improved CPU performance. | | [[https://github.com/ikawrakow/ik_llama.cpp|ik_llama]] | Improved CPU performance. |
  
-=== Engine Manangers ===+==== Engine Manangers ====
 These tools are designed to provide a user-friendly layer which handles backend needs and engine management simultaneously. If you're not sure what to pick, this is a good place to start. These tools are designed to provide a user-friendly layer which handles backend needs and engine management simultaneously. If you're not sure what to pick, this is a good place to start.
  
 ^ Name ^ Notes ^ ^ Name ^ Notes ^
-| [[https://github.com/oobabooga/text-generation-webui|text-generation-webui (Oobabooga)]] | Most popular backend, for good reason. | +|  **Oft-Recommended Managers**  || 
-| [[https://github.com/LostRuins/koboldcpp|koboldcpp]] | Good UI, capable of managing itself. |+| [[https://github.com/oobabooga/text-generation-webui|text-generation-webui (Oobabooga)]] | Offers all other engines here and more. | 
 +| [[https://github.com/LostRuins/koboldcpp|koboldcpp]] | Good UI, koboldcpp engine only
 +|  **Other Managers**  ||
 | [[https://localai.io/|LocalAI]] | Provides OpenAI-compatible API. | | [[https://localai.io/|LocalAI]] | Provides OpenAI-compatible API. |
 | [[https://ollama.com/|ollama]] | Wraps llama.cpp. | | [[https://ollama.com/|ollama]] | Wraps llama.cpp. |
 | [[https://github.com/theroyallab/tabbyAPI|tabbyAPI]] | Official API server for ExLlama engines. | | [[https://github.com/theroyallab/tabbyAPI|tabbyAPI]] | Official API server for ExLlama engines. |
- 
- 
-**Note**: This list does not include backends that can directly accept Character Cards - those are considered [[interfaces]]. 
  
 ===== 3rd-Party API providers ===== ===== 3rd-Party API providers =====
-These are essentially remote backends. Everything you send and receive is, at minimum, available to the provider(s). Censorship can be expected, to varying degrees.+These are essentially remote backends. Everything you send and receive is, at minimum, available to the provider(s). Censorship is often encountered to varying degrees.
  
  
Line 45: Line 46:
  
 Additionally, most commercial APIs can be utilized such as ChatGPT, Claude, Perplexity, etc. Additionally, most commercial APIs can be utilized such as ChatGPT, Claude, Perplexity, etc.
 +
 +===== Additional Resources =====
 +SillyTavern's page on [[https://docs.sillytavern.app/usage/api-connections/|API Connections]].
backends.1773765296.txt.gz · Last modified: by tys