[Rust Frontend] Pass vision preprocessing context for Nemotron-H (#57634)

Pass the remaining engine context-length budget to model-owned vision processors through VisionPreprocessingContext. Preserve Nemotron batched engine fields and recognize llm_config as a text_config alias.

Use the merged upstream llm-multimodal revision f0985ef65967615db2c79279aa07818499301bfd.

Co-authored-by: Codex <noreply@openai.com>
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
This commit is contained in:
Biswa Panda
2026-09-24 14:08:26 +08:00
committed by GitHub
co-authored by Codex
parent ee2f3cdf12
commit cf78797cb7
8 changed files with 240 additions and 21 deletions
+1 -1
View File
@@ -61,7 +61,7 @@ indexmap = "2.13.0"
indicatif = "0.18.4"
itertools = "0.14.0"
libc = "0.2.177"
llm-multimodal = { git = "https://github.com/smg-project/llm-multimodal", rev = "53ca0d6ea829af7477b6b63851cc70bed409bbc8", default-features = false, features = ["native-tls"] }
llm-multimodal = { git = "https://github.com/smg-project/llm-multimodal", rev = "f0985ef65967615db2c79279aa07818499301bfd", default-features = false, features = ["native-tls"] }
mimalloc = "0.1.52"
minijinja = { version = "2.24", features = ["unstable_machinery", "json", "builtins", "loader", "loop_controls", "preserve_order"] }
minijinja-contrib = { version = "2.24", features = ["pycompat"] }