v0.5.9 - merge openai api into main loop

allows openai api to output the same image/speech as candle does!

TODO:
may have spacing issues at times, the output fromt the the stream
output from the web api of llama.cpp seems to contain some empty
new line only tokens we aren't handling perfectly which leads
to some concatenated sentence end/beginnings.

Cargo.toml

@@ -9,7 +9,7 @@ license-file = "LICENSE"
 homepage = "https://github.com/groovybits/rsllm/wiki"
 repository = "https://github.com/groovybits/rsllm"
 authors = ["Chris Kennedy"]
-version = "0.5.8"
+version = "0.5.9"
 edition = "2021"
 
 [lib]

scripts/twitch.sh

@@ -18,8 +18,8 @@ ALIGNMENT=right
 TEMPERATURE=0.8
 CONTINUOUS=1
 POLL_INTERVAL=0
-PIPELINE_CONCURRENCY=8
-TWITCH_LLM_CONCURRENCY=2
+PIPELINE_CONCURRENCY=1
+TWITCH_LLM_CONCURRENCY=1
 CONTEXT_SIZE=4000
 SUBTITLES=1
 DAEMON=1
@@ -29,14 +29,19 @@ GREETING="Hi I'm Alice, ask me a question!"
 TWITCH_CHAT_HISTORY=8
 TWITCH_MODEL=mistral
 ASYNC_CONCURRENCY=0
+USE_API=1
 # === END OF CONFIGURATION ===
 #
 #
+USE_API_CMD=
 SUBTITLE_CMD=
 DAEMON_CMD=
 NO_HISTORY_CMD=
 QUANTIZED_CMD=
 ASYNC_CONCURRENCY_CMD=
+if [ "$USE_API" == 1 ]; then
+    USE_API_CMD="--use-api"
+fi
 if [ "$ASYNC_CONCURRENCY" == 1 ]; then
     ASYNC_CONCURRENCY_CMD="--async-concurrency"
 fi
@@ -81,6 +86,7 @@ DYLD_LIBRARY_PATH=`pwd`:/usr/local/lib:$DYLD_LIBRARY_PATH \
     --llm-history-size $CONTEXT_SIZE \
     --greeting "$GREETING" \
     $SINGLE_CONCURRENCY_CMD \
+    $USE_API_CMD \
     $SUBTITLE_CMD \
     $DAEMON_CMD \
     $CONTINUOUS_CMD \

src/args.rs

@@ -4,7 +4,7 @@ use clap::Parser;
 #[derive(Parser, Debug, Clone)]
 #[clap(
     author = "Chris Kennedy",
-    version = "0.5.8",
+    version = "0.5.9",
     about = "Rust AI Stream Analyzer Twitch Bot"
 )]
 pub struct Args {