Blog
Replicate
STABLE

LLama 2 13B Chat GGUF

Llama-2 13B chat with support for grammars and jsonschema

Inputs
Prompt
Prompt
Frequency Penalty
Frequency penalty
Grammar
Grammar in GBNF format. Use either grammar or jsonschema.
Jsonschema
JSON schema for the generated output. Use either grammar or jsonschema. You can use the jsonschema in the prompt by using the special string '{jsonschema}'
Max Tokens
Max number of tokens to return
Mirostat Entropy
Mirostat target entropy
Mirostat Learning Rate
Mirostat learning rate, if mirostat_mode is not Disabled
mirostat_mode
Mirostat sampling mode
Presence Penalty
Presence
Repeat Penalty
Repetition penalty
Temperature
Temperature
Top K
Top K
Top P
Top P
Outputs
Json Data
JSON data (array or object)