chore: update Dockerfile, TinyLlama converter, vite config for Space

Dockerfile

@@ -48,7 +48,7 @@ RUN echo 'server { \
     location /api/hf-inference/ { \
         resolver 127.0.0.11 8.8.8.8 valid=30s; \
         set $hf_upstream https://router.huggingface.co; \
-        rewrite ^/api/hf-inference/(.*)$ /hf-inference/$1 break; \
+        rewrite ^/api/hf-inference/(.*)$ /$1 break; \
         proxy_pass $hf_upstream; \
         proxy_http_version 1.1; \
         proxy_set_header Host router.huggingface.co; \

src/components/TinyLlamaConverter.jsx

@@ -46,7 +46,7 @@ const TinyLlamaConverter = ({ addToMessage, sharedInput, setSharedInput }) => {
     try {
       if (config.provider === 'hf') {
         const model = config.model ?? HF_MODEL;
-        const url = `${config.baseUrl}/models/${model}/v1/chat/completions`;
+        const url = `${config.baseUrl}/v1/chat/completions`;
         const res = await fetch(url, {
           method: 'POST',
           headers: { 'Content-Type': 'application/json' },

vite.config.js

@@ -20,7 +20,7 @@ export default defineConfig({
       '/api/hf-inference': {
         target: 'https://router.huggingface.co',
         changeOrigin: true,
-        rewrite: (path) => path.replace(/^\/api\/hf-inference/, '/hf-inference'),
+        rewrite: (path) => path.replace(/^\/api\/hf-inference/, ''),
         configure: (proxy) => {
           const token = process.env.VITE_HF_TOKEN;
           if (token) {