mtmd, server: add "placeholder bitmap" for counting tokens , add */input_tokens API (#23913)
* mtmd: add "placeholder bitmap" for counting tokens w/o preprocessing * fast path skip preproc for placeholder * fix build * correct the api * add server endpoint + tests * add object name * update docs * add proxy handling * fix build * fix audio input path * use is_placeholder in process_mtmd_prompt() * nits * nits (2) * docs: clarify chat/completions/input_tokens is not official * fix merge problem
This commit is contained in:
@@ -110,8 +110,10 @@ struct server_routes {
|
||||
server_http_context::handler_t post_completions;
|
||||
server_http_context::handler_t post_completions_oai;
|
||||
server_http_context::handler_t post_chat_completions;
|
||||
server_http_context::handler_t post_chat_completions_tok;
|
||||
server_http_context::handler_t post_control;
|
||||
server_http_context::handler_t post_responses_oai;
|
||||
server_http_context::handler_t post_responses_tok_oai;
|
||||
server_http_context::handler_t post_transcriptions_oai;
|
||||
server_http_context::handler_t post_anthropic_messages;
|
||||
server_http_context::handler_t post_anthropic_count_tokens;
|
||||
@@ -139,6 +141,7 @@ private:
|
||||
std::unique_ptr<server_res_generator> handle_slots_restore(const server_http_req & req, int id_slot);
|
||||
std::unique_ptr<server_res_generator> handle_slots_erase(const server_http_req &, int id_slot);
|
||||
std::unique_ptr<server_res_generator> handle_embeddings_impl(const server_http_req & req, task_response_type res_type);
|
||||
std::unique_ptr<server_res_generator> handle_count_tokens(const llama_vocab * vocab, mtmd_context * mctx, const server_http_req & req, task_response_type res_type);
|
||||
|
||||
// using unique_ptr to allow late initialization of const
|
||||
std::unique_ptr<const server_context_meta> meta;
|
||||
|
||||
Reference in New Issue
Block a user