Unverified50% confidenceFactExact time
OpenAI's Chat Completions API offers the logit_bias parameter, which accepts a dictionary mapping token IDs to bias values ranging from -100 to 100, with negative values suppressing and positive values boosting.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedOpenAI released the public beta of the Decisions API, with the endpoint POST /v1/decisions, which returns only typed answers (probability, choice, or score).66% similarUnverifiedOpenAI is publicly testing the Decisions API, powered by an internal frontier model, with decision speeds up to 10 times faster than flagship models, supporting text and image input, and outputting in three forms: probabilities, choices, and scores.63% similarUnverifiedThe scoring API can rate the relevance/quality of retrieved candidate documents or candidate answers generated by a model, helping to select the optimal results to reduce hallucinations.62% similarUnverifiedThe author recommends enabling mandatory checksum verification by default for symbologies that support check digits, and making round-trip success the return condition at the API layer.61% similarUnverifiedSpeculative decoding can now report per-request acceptance stats via the API.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1067602API
curl https://kongchang.com/api/v1/knowledge/claims/1067602MCP
get_claim(id=1067602)