spec: single device drafter should create meta backend wrapper - #28390
Merged
Conversation
ggerganov
reviewed
Sep 4, 2026
Contributor
|
As discussed, I added the small change that draft devices defaults to global devices value, this is way more intuitive and avoids having to specify them separately in many cases. |
am17an
commented
Sep 6, 2026
| | `--reasoning-budget N` | token budget for thinking: -1 for unrestricted, 0 for immediate end, N>0 for token budget (default: -1)<br/>(env: LLAMA_ARG_THINK_BUDGET) | | ||
| | `--reasoning-budget-message MESSAGE` | message injected before the end-of-thinking tag when reasoning budget is exhausted (default: none)<br/>(env: LLAMA_ARG_THINK_BUDGET_MESSAGE) | | ||
| | `--reasoning-preserve, --no-reasoning-preserve` | preserve reasoning trace in the full history, not just the last assistant message (default: template default)<br/>compatible with certain templates having 'supports_preserve_reasoning' capability<br/>example: https://docs.z.ai/guides/capabilities/thinking-mode#preserved-thinking<br/>(env: LLAMA_ARG_REASONING_PRESERVE) | | ||
| | `--reasoning-preserve, --no-reasoning-preserve` | preserve reasoning trace in the full history, not just the last assistant message (default: enabled)<br/>compatible with certain templates having 'supports_preserve_reasoning' capability<br/>example: https://docs.z.ai/guides/capabilities/thinking-mode#preserved-thinking<br/>(env: LLAMA_ARG_REASONING_PRESERVE) | |
Contributor
Author
There was a problem hiding this comment.
probably needs a rebase
Contributor
There was a problem hiding this comment.
Did it change again? Otherwise this just means someone forgot to regenerate the readmes when changing the default for reasoning-preserve.
ggerganov
approved these changes
Sep 6, 2026
github-actions
Bot
added
documentation
ruixiang63
approved these changes
Sep 8, 2026
x1250
pushed a commit
to x1250/llama.cpp
that referenced
this pull request
Sep 9, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Overview
when using
--spec-draft-device CUDA0but also-sm tensor, we will create a Meta backend context which consumes some VRAM which is not necessary as it's never used. This PR adds a special case to stop that from happeningAdditional information
Requirements