PulseAugur
EN
LIVE 08:22:24

llama.cpp server accepts null sampling parameters, aligning with OpenAI spec

By PulseAugur Editorial · [1 sources] ·

The llama.cpp project has released version b9967, introducing an update to its server functionality. This update allows the server to accept null sampling parameters, such as temperature and top_p. By treating null values as absent, clients can now send null for these parameters to utilize the server's default settings, aligning with OpenAI's specifications and existing conventions. AI

IMPACT Minor improvement for developers using llama.cpp, enhancing compatibility with OpenAI's parameter handling.

RANK_REASON This is a software update for a specific tool, not a frontier release or significant industry event.

Read on llama.cpp — Releases →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

llama.cpp server accepts null sampling parameters, aligning with OpenAI spec

COVERAGE [1]

  1. llama.cpp — Releases TIER_1 English(EN) · ServeurpersoCom ·

    b9967: server: accept null sampling params (#25538)

    <ul> <li>server: accept null sampling params</li> </ul> <p>Extend the schema validation to treat a null value as absent, so<br /> clients can send null on nullable params (temperature, top_p, ...)<br /> to request the server default. This matches the OpenAI spec and the<br /> jso…

RELATED ENTITIES

RELATED TOPICS