{{Guidance-ai}} API

Question 1

{{Guidance}}

Answer 1

Guidance enables you to control modern language models more effectively and efficiently than traditional prompting or chaining. Guidance programs allow you to interleave generation, prompting, and logical control into a single continuous flow matching how the language model actually processes the text.

Question 2

guidance.library.add(*args)

Answer 2

Add the given variables together.

Question 3

guidance.library.assistant([hidden,…])

Answer 3

A chat role block for the ‘assistant’ role.

Question 4

guidance.library.await_(name[,consume,…])

Answer 4

Awaits a variable by returning its value and then deleting it.

Question 5

guidance.library.block([name,hidden,…])

Answer 5

Generic block-level element.

Question 6

guidance.library.break_()

Answer 6

Breaks out of the current loop.

Question 7

guidance.library.each(list[,hidden,…])

Answer 7

Iterate over a list and execute a block for each item.

Question 8

guidance.library.equal(*args)

Answer 8

Check that all arguments are equal.

Question 9

guidance.library.gen([name,stop,…])

Answer 9

Use the LLM to generate a completion.

Question 10

guidance.library.geneach(list_name[,stop,…])

Answer 10

Generate a potentially variable length list of items using the LLM.

Question 11

guidance.library.if_(value,*[,invert,…])

Answer 11

Standard if/else statement.

Question 12

guidance.library.role(role_name[,hidden,…])

Answer 12

A chat role block.

Question 13

guidance.library.select([variable_name,…])

Answer 13

Select a value from a list of choices.

Question 14

guidance.library.set(name[,value,hidden,…])

Answer 14

Set the value of a variable or set of variables.

Question 15

guidance.library.shell(command[,safe,…])

Answer 15

Execute a shell command on the local machine (with user confirmation by default).

Question 16

guidance.library.strip(string)

Answer 16

Strip whitespace from the beginning and end of the given string.

Question 17

guidance.library.subtract(minuend,subtrahend)

Answer 17

Subtract the second variable from the first.

Question 18

guidance.library.system([hidden,…])

Answer 18

A chat role block for the ‘system’ role.

Question 19

guidance.library.unless(value[,_parser_context])

Answer 19

[blank in the documentation]

Question 20

guidance.library.user([hidden,_parser_context])

Answer 20

A chat role block for the ‘user’ role.

Question 21

guidance.llms.OpenAI([model,caching,…])

Answer 21

[blank in the documentation]

Question 22

guidance.llms.Transformers([model,…])

Answer 22

A HuggingFace transformers language model with Guidance support.

Question 23

guidance.llms.transformers.LLaMA([model,…])

Answer 23

A HuggingFace transformers version of the LLaMA language model with Guidance support.

Question 24

guidance.llms.transformers.MPT([model,…])

Answer 24

A HuggingFace transformers version of the MosaicML MPT language model with Guidance support.

Question 25

guidance.llms.transformers.Vicuna([model,…])

Answer 25

A HuggingFace transformers version of the Vicuna language model with Guidance support.

Question 26

guidance.llms.transformers.Koala([model,…])

Answer 26

A HuggingFace transformers version of the Koala language model with Guidance support.