feat: auto fit tensors across devices to guarantee optimal load (#1736)

This commit is contained in:
Piotr Wilkin (ilintar)
2026-07-04 18:48:10 +02:00
committed by GitHub
parent 68f3d6df9f
commit b11c95a41c
8 changed files with 493 additions and 16 deletions

View File

@@ -228,6 +228,7 @@ typedef struct {
const char* backend;
const char* params_backend;
const char* split_mode; // weight distribution for multi-device modules: layer (default) or row, or per-module assignments e.g. "diffusion=row"
bool auto_fit;
const char* rpc_servers;
} sd_ctx_params_t;