Clean up code and rename sanity-checks

Author: akaashrp

.gitignore

@@ -324,5 +324,4 @@ node_modules
 lib
 .parcel-cache
 
-examples/tests
 **/.next

examples/get-started-latency-breakdown/src/get_started_latency_breakdown.ts

@@ -45,7 +45,6 @@ async function main() {
     setLabel("init-label", report.text);
   };
   // Option 1: If we do not specify appConfig, we use `prebuiltAppConfig` defined in `config.ts`
-  // const selectedModel = "Llama-3.1-8B-Instruct-q4f32_1-MLC";
   const selectedModel = "Qwen3-0.6B-q0f32-MLC";
   const engine: webllm.MLCEngineInterface = await webllm.CreateMLCEngine(
     selectedModel,
@@ -61,34 +60,6 @@ async function main() {
     },
   );
 
-  // Option 2: Specify your own model other than the prebuilt ones
-  // const appConfig: webllm.AppConfig = {
-  //   model_list: [
-  //     {
-  //       model: "https://huggingface.co/mlc-ai/Llama-3.1-8B-Instruct-q4f32_1-MLC",
-  //       model_id: "Llama-3.1-8B-Instruct-q4f32_1-MLC",
-  //       model_lib:
-  //         webllm.modelLibURLPrefix +
-  //         webllm.modelVersion +
-  //         "/Llama-3_1-8B-Instruct-q4f32_1-ctx4k_cs1k-webgpu.wasm",
-  //       overrides: {
-  //         context_window_size: 2048,
-  //       },
-  //     },
-  //   ],
-  // };
-  // const engine: webllm.MLCEngineInterface = await webllm.CreateMLCEngine(
-  //   selectedModel,
-  //   { appConfig: appConfig, initProgressCallback: initProgressCallback },
-  // );
-
-  // Option 3: Instantiate MLCEngine() and call reload() separately
-  // const engine: webllm.MLCEngineInterface = new webllm.MLCEngine({
-  //   appConfig: appConfig, // if do not specify, we use webllm.prebuiltAppConfig
-  //   initProgressCallback: initProgressCallback,
-  // });
-  // await engine.reload(selectedModel);
-
   const latencyBreakdown: LatencyBreakdown = {
     logitProcessorTime: [],
     logitBiasTime: [],
@@ -97,7 +68,7 @@ async function main() {
     totalTime: [],
     grammarBitmaskTime: [],
   };
-  // want decode_tokens_per_s, e2e_latency_s, time_per_output_token_s, completion_tokens
+
   const decodeTokensPerS: number[] = [];
   const completionTokens: number[] = [];
   const e2eLatencyS: number[] = [];

examples/sanity-checks/README.md …es/tests/scripts/sanity_checks/README.mdexamples/sanity-checks/README.md renamed to examples/tests/scripts/sanity_checks/README.md examples/sanity-checks/README.md renamed to examples/tests/scripts/sanity_checks/README.md

@@ -1,10 +1,10 @@
 {
-  "name": "sanity-checks",
+  "name": "tests",
   "version": "0.1.0",
   "private": true,
   "scripts": {
-    "start": "parcel src/sanity_checks.html --port 8889",
-    "build": "parcel build src/sanity_checks.html --dist-dir lib"
+    "start": "parcel sanity_checks.html --port 8889",
+    "build": "parcel build sanity_checks.html --dist-dir lib"
   },
   "devDependencies": {
     "buffer": "^5.7.1",

examples/sanity-checks/package.json …tests/scripts/sanity_checks/package.jsonexamples/sanity-checks/package.json renamed to examples/tests/scripts/sanity_checks/package.json examples/sanity-checks/package.json renamed to examples/tests/scripts/sanity_checks/package.json

@@ -71,7 +71,6 @@ async function testLogitProcessor(
 async function testLogitBias(modelId: string, appConfig: webllm.AppConfig) {
   // Set logit_bias to strongly favor token 0
   const prompt = "Test logit bias.";
-  // const t0 = performance.now();
   const engine: webllm.MLCEngineInterface = await createEngine(
     modelId,
     appConfig,
@@ -128,7 +127,6 @@ async function testPenalties(modelId: string, appConfig: webllm.AppConfig) {
 async function testLogprobs(modelId: string, appConfig: webllm.AppConfig) {
   // Test logprobs: check that logprobs are returned and sum to ~1 after exp
   const prompt = "Test logprobs.";
-  const t0 = performance.now();
   const engine: webllm.MLCEngineInterface = await createEngine(
     modelId,
     appConfig,
@@ -140,7 +138,6 @@ async function testLogprobs(modelId: string, appConfig: webllm.AppConfig) {
     logprobs: true,
     top_logprobs: 5,
   });
-  const t1 = performance.now();
   const logprobs = reply.choices[0]?.logprobs;
 
   let logprobsAllCloseTo1 = true;