[Example] ggml: add phi-3-mini test (#136)

Signed-off-by: dm4 <[email protected]>

.github/workflows/llama.yml

@@ -259,6 +259,18 @@ jobs:
                 default \
                 $'[INST] <<SYS>>\nYou are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.  Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you do not know the answer to a question, please do not share false information.\n<</SYS>>\nWhat is the capital of Japan?[/INST]'
 
+          - name: Phi 3 Mini
+            run: |
+              test -f ~/.wasmedge/env && source ~/.wasmedge/env
+              cd wasmedge-ggml/test/phi-3-mini
+              curl -LO https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-gguf/resolve/main/Phi-3-mini-4k-instruct-q4.gguf
+              cargo build --target wasm32-wasi --release
+              time wasmedge --dir .:. \
+                --nn-preload default:GGML:AUTO:Phi-3-mini-4k-instruct-q4.gguf \
+                target/wasm32-wasi/release/wasmedge-ggml-phi-3-mini.wasm \
+                default \
+                $'<|user|>\nWhat is the capital of Japan?<|end|>\n<|assistant|>'
+
           - name: Build llama-stream
             run: |
               cd wasmedge-ggml/llama-stream

wasmedge-ggml/test/phi-3-mini/Cargo.toml

@@ -0,0 +1,8 @@
+[package]
+name = "wasmedge-ggml-phi-3-mini"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+serde_json = "1.0"
+wasmedge-wasi-nn = "0.7.1"

wasmedge-ggml/test/phi-3-mini/README.md

@@ -0,0 +1,13 @@
+# `phi-3-mini`
+
+Ensure that we can use the `phi-3-mini` model.
+
+## Execute
+
+```console
+$ curl -LO https://huggingface.co/microsoft/Phi-3-mini-4k-instruct-gguf/resolve/main/Phi-3-mini-4k-instruct-q4.gguf
+$ wasmedge --dir .:. \
+  --nn-preload default:GGML:AUTO:Phi-3-mini-4k-instruct-q4.gguf \
+  wasmedge-ggml-phi-3-mini.wasm default \
+  <prompt>
+```

wasmedge-ggml/test/phi-3-mini/src/main.rs

@@ -0,0 +1,97 @@
+use serde_json::json;
+use serde_json::Value;
+use std::env;
+use wasmedge_wasi_nn::{
+    self, BackendError, Error, ExecutionTarget, GraphBuilder, GraphEncoding, GraphExecutionContext,
+    TensorType,
+};
+
+fn get_options_from_env() -> Value {
+    let mut options = json!({});
+    if let Ok(val) = env::var("enable_log") {
+        options["enable-log"] = serde_json::from_str(val.as_str())
+            .expect("invalid value for enable-log option (true/false)")
+    } else {
+        options["enable-log"] = serde_json::from_str("false").unwrap()
+    }
+    if let Ok(val) = env::var("n_gpu_layers") {
+        options["n-gpu-layers"] =
+            serde_json::from_str(val.as_str()).expect("invalid ngl value (unsigned integer")
+    } else {
+        options["n-gpu-layers"] = serde_json::from_str("0").unwrap()
+    }
+
+    options
+}
+
+fn get_data_from_context(context: &GraphExecutionContext, index: usize) -> String {
+    // Preserve for 4096 tokens with average token length 6
+    const MAX_OUTPUT_BUFFER_SIZE: usize = 4096 * 6;
+    let mut output_buffer = vec![0u8; MAX_OUTPUT_BUFFER_SIZE];
+    let mut output_size = context
+        .get_output(index, &mut output_buffer)
+        .expect("Failed to get output");
+    output_size = std::cmp::min(MAX_OUTPUT_BUFFER_SIZE, output_size);
+
+    return String::from_utf8_lossy(&output_buffer[..output_size]).to_string();
+}
+
+fn get_output_from_context(context: &GraphExecutionContext) -> String {
+    get_data_from_context(context, 0)
+}
+
+fn main() {
+    let args: Vec<String> = env::args().collect();
+    let model_name: &str = &args[1];
+
+    // Set options for the graph. Check our README for more details:
+    // https://github.com/second-state/WasmEdge-WASINN-examples/tree/master/wasmedge-ggml#parameters
+    let options = get_options_from_env();
+
+    // This is mainly for the CI workflow. Only support the prompt from the command line.
+    if args.len() < 3 {
+        println!("Usage: {} <model_name> <prompt>", args[0]);
+        std::process::exit(1);
+    }
+    let prompt = &args[2];
+
+    // Create graph and initialize context.
+    let graph = GraphBuilder::new(GraphEncoding::Ggml, ExecutionTarget::AUTO)
+        .config(serde_json::to_string(&options).expect("Failed to serialize options"))
+        .build_from_cache(model_name)
+        .expect("Failed to build graph");
+    println!("Graph {} loaded.", graph);
+    let mut context = graph
+        .init_execution_context()
+        .expect("Failed to init context");
+
+    // Set the prompt.
+    println!("Prompt:\n{}", prompt);
+    let tensor_data = prompt.as_bytes().to_vec();
+    context
+        .set_input(0, TensorType::U8, &[1], &tensor_data)
+        .expect("Failed to set input");
+    println!("Response:");
+
+    // Execute the inference.
+    match context.compute() {
+        Ok(_) => (),
+        Err(Error::BackendError(BackendError::ContextFull)) => {
+            println!("\n[INFO] Context full, we'll reset the context and continue.");
+        }
+        Err(Error::BackendError(BackendError::PromptTooLong)) => {
+            println!("\n[INFO] Prompt too long, we'll reset the context and continue.");
+        }
+        Err(err) => {
+            println!("\n[ERROR] {}", err);
+            std::process::exit(1);
+        }
+    }
+
+    // Retrieve the output.
+    let output = get_output_from_context(&context);
+    println!("{}", output.trim());
+
+    // Unload.
+    graph.unload().expect("Failed to unload graph");
+}