# llama.cpp adds logprob support for speculative decoding

> The open‑source Llama.cpp library now returns token log probabilities when using speculative decoding, fixing a regression on the master branch.

Oossa · 2026-10-21 · https://oossa.com/en/llama-cpp-adds-logprob-support-for-speculative-decoding

The ggml‑org/llama.cpp project released version b11776 on 2026-10-21. The update adds support for returning log probabilities (logprobs) while running speculative decoding, a technique that speeds up generation by guessing tokens ahead of time. A regression test was also added to ensure the new logprob values match a run without speculation. The fix resolves a bug where the master branch returned empty logprob data.

## The facts

- Version b11776 adds logprob output for speculative decoding (#27196).
- Release date: Wed Oct 21 2026.

## Why it matters

Developers can now see confidence scores for each token when using fast speculative decoding, helping them debug and improve generation quality.

## Sources & references

1. [ggml-org/llama.cpp b11776](https://github.com/ggml-org/llama.cpp/releases/tag/b11776) – llama.cpp, 2026-10-21

Last updated: 2026-10-21