releasesggml/llama.cpp releasesJul 29, 2026
ggml/llama.cpp releases: b10182
Sentiment: neutral
TL;DR
The ggml/llama.cpp project released a new version addressing security issues by moving suppress_tokens handling to common/sampling and removing has_logit_bias, emphasizing improved safety in the latest update.
Detailed Summary
The ggml/llama.cpp project released an update addressing security issues by moving suppress_tokens handling to common/sampling (#26276). This update also removed has_logit_bias. The broader impact includes enhancing the security and functionality of the llama model, benefiting users across various platforms including macOS and iOS.
Key Points
- • llama: move suppress_tokens handling to common/sampling
- • address security issues
- • rm has_logit_bias