Add vocab_only

[ardinursyamsu]

Would you add vocab_only option just like python library do? I want to separate my tokenizer service and model service in my spring boot app.

[kherud]

Hey @ardinursyamsu past versions of this Java binding allowed this too, so it shouldn't be too much effort to bring this back. I'll have a look over the next few days.

[bernardladenthin]

I noticed the repository hasn't seen recent activity, so I've taken the liberty of forking it and continuing development at: https://github.com/bernardladenthin/java-llama.cpp

See also #125

[bernardladenthin]

Implemented in the downstream fork at https://github.com/bernardladenthin/java-llama.cpp.
ModelParameters.setVocabOnly() (ModelParameters.java:1336) toggles
ModelFlag.VOCAB_ONLY → --vocab-only (args/ModelFlag.java:92); the
native layer branches on jctx->vocab_only to skip server/worker init while
keeping the vocab available for tokenize/decode
(src/main/cpp/jllama.cpp:710-718, 923, 944).