Skip to content

Compute buffer and KV-cache aware layer distribution for multi-GPU inference #14484

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
borebot wants to merge 2 commits into
base: master
Choose a base branch
from
+316 −0
Open

Compute buffer and KV-cache aware layer distribution for multi-GPU inference #14484

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
49f9271
Implement context-length dependent KV-cache and Compute Buffer aware …
borebot Jul 1, 2025
aa89dda
Merge branch 'master' into kv-compute-buffer-cache-aware-allocation
borebot Feb 3, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions common/common.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1367,6 +1367,7 @@ struct llama_model_params common_model_params_to_llama(common_params & params) {
mparams.use_direct_io = params.use_direct_io;
mparams.use_mlock = params.use_mlock;
mparams.check_tensors = params.check_tensors;
mparams.requested_n_ctx = params.n_ctx;
mparams.use_extra_bufts = !params.no_extra_bufts;
mparams.no_host = params.no_host;

Expand Down
3 changes: 3 additions & 0 deletions include/llama.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -306,6 +306,9 @@ extern "C" {
// override key-value pairs of the model meta data
const struct llama_model_kv_override * kv_overrides;

// expected context size for memory allocation planning (0 = auto)
uint32_t requested_n_ctx;

// Keep the booleans together to avoid misalignment during copy-by-value.
bool vocab_only; // only load the vocabulary, no weights
bool use_mmap; // use mmap if possible
Expand Down
Loading