Skip to main content

Chat Completions

ParameterWhat does this do?TelnyxOpenAI
messagesProvides chat context
modelAdjusts speed + quality
streamStreams response
max_tokensLimits output length
temperatureAdjusts predictability
top_pAdjusts variety
frequency_penaltyDecreases repetition
presence_penaltyDecreases repetition...
nReturns n responses
stopForces model to stop
logit_biasTweaks odds of results
logprobsReturns odds of outputs
-> top_logprobs-> For how many candidates?
seedReduces randomness
response_formatEnsures syntax (e.g. JSON)
guided_jsonEnsures output conforms to schema
guided_regexEnsures output conforms to regex
guided_choiceEnsures output conforms to multiple choice
min_ptop_p alternative
use_beam_searchExplores more options
-> best_of-> How many options?
-> length_penalty-> Are long options bad?
-> early_stopping-> How hard should it try?
toolsHelps model respond
-> functions-> Outputs JSON for your code🔜
-> retrieval-> Uses your docs (e.g. PDFs)
tool_choiceHow does model choose?🔜
userTracks users

Transcriptions (BETA)

ParameterWhat does this do?TelnyxOpenAI
fileProvides audio data
modelAdjusts speed + quality
response_formatAdjusts output format
timestamp_granularities[]Adds timestamps
-> segment-> per audio segment
-> word-> per word
languageImproves accuracy
promptGuides style
temperatureAdjusts "creativity"