Coding Tools
llama.cpp fixes server crash from long token names
llama.cpp Releaseshigh confidence
Why it matters
- →Fixes a denial-of-service vector where long token names crash the llama-server process.
- →Removes dependency on recursive regex for special token escaping, improving stability.
- →Applies to all binary distributions without requiring code changes from users.
llama.cpp has released a fix for a server-side crash triggered by long special tokens in request text. The issue stemmed from libstdc++'s std::regex_replace implementation, which recursed on each character of the token name, causing a stack overflow when names exceeded approximately 30 characters. The fix replaces the regex call with a manual loop, resolving the segmentation fault while maintaining compatibility across all supported platforms including CUDA, ROCm, and Apple Silicon builds.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Llama.cpp b9870 Release Fixes Long Reasoning Loops
llama.cpp Releases · July 4, 2026 · Same story
What happened next
- 2llama.cpp fixes grammar rejection for long tokens
llama.cpp Releases · October 26, 2026 · Same story
Follow this story
Open the full story →More from llama.cpp Releases
Coding Toolscoding
llama.cpp b12022 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing ROCm 10.0 to Linux and Windows alongside CUDA 13.4, effectively closing the hardware gap for AMD users who were previously left behind. The inclusion of native Snapdragon support on Linux marks a significant step toward running local LLMs on ARM-based mobile chips without emulation overhead. While KleidiAI on Apple Silicon is disabled in this build, the broader expansion to diverse accelerators ensures that developers aren't locked into NVIDIA ecosystems anymore.
llama.cpp Releases
Coding Toolscoding
llama.cpp b12023 fixes shared prefix bug
This release addresses a specific inference artifact where decision prefixes were incorrectly bleeding into output embeddings. By capping shared prefixes at the first output position, llama.cpp ensures that option tokens remain distinct for parent and child tasks. This is a targeted fix for multi-step reasoning or voting workflows rather than a performance boost. It stabilizes complex generation patterns without changing the underlying model architecture.
More in Coding Tools
Coding Toolscoding
Claude Code v2.1.316: Subagents and Gateway Budgets
Anthropic is tightening control over Claude Code's autonomous behavior with new environment variables that let you disable 'ultrathink' or cap subagent effort levels. For teams, the introduction of pooled group budgets on the gateway allows shared monthly spending limits across members, a practical fix for enterprise cost management. The release also patches critical bugs where shell commands would hang or crash on exit, stabilizing the developer experience significantly.
Claude Code Releases
© The Verge AICoding Toolscoding
