Skip to content

Commit

Permalink
Merge pull request #188 from alabulei1/alabulei1-patch-1
Browse files Browse the repository at this point in the history 
Signed-off-by: Michael Yuan <[email protected]>
  • Loading branch information
@juntao
juntao authored Nov 6, 2023
2 parents e3811ce + 6036292 commit 514afc1
Show file tree
Hide file tree
Showing 2 changed files with 185 additions and 120 deletions.
170 changes: 100 additions & 70 deletions docs/develop/rust/wasinn/llm_inference.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,35 +13,36 @@ WasmEdge now supports Llama2, Codellama-instruct, BELLE-Llama, Mistral-7b-instru
Besides the [regular WasmEdge and Rust requirements](../../rust/setup.md), please make sure that you have the [Wasi-NN plugin with ggml installed](../../../start/install.md#wasi-nn-plug-in-with-ggml-backend).

## Quick start

Because the example already includes a compiled WASM file from the Rust code, we could use WasmEdge CLI to execute the example directly. First, git clone the `llama-utils` repo.

```bash
git clone https://github.com/second-state/llama-utils.git
cd chat
curl -LO https://github.com/second-state/llama-utils/raw/main/chat/llama-chat.wasm
```

Next, let's get the model. In this example, we are going to use the llama2 7b chat model in GGUF format. You can also use other kinds of llama2 models, check out [here](https://github.com/second-state/llama-utils/blob/main/chat/README.md#get-model).

```bash
git clone curl -LO https://huggingface.co/wasmedge/llama2/blob/main/llama-2-7b-chat-q5_k_m.gguf
curl -LO https://huggingface.co/wasmedge/llama2/blob/main/llama-2-7b-chat-q5_k_m.gguf
```

Run the inference application in WasmEdge.

```bash
wasmedge --dir .:. \
--nn-preload default:GGML:CPU:llama-2-7b.Q5_K_M.gguf llama-chat.wasm default \
--prompt 'Robert Oppenheimer most important achievement is ' \
--ctx-size 4096
wasmedge --dir .:. --nn-preload default:GGML:AUTO:llama-2-7b-chat-q5_k_m.gguf llama-chat.wasm
```

After executing the command, you may need to wait a moment for the input prompt to appear. Once the execution is complete, the following output will be generated.
After executing the command, you may need to wait a moment for the input prompt to appear. You can enter your question once you see the `[USER]:` prompt:

```bash
Robert Oppenheimer most important achievement is
1945 Manhattan Project.
Robert Oppenheimer was born in New York City on April 22, 1904. He was the son of Julius Oppenheimer, a wealthy German-Jewish textile merchant, and Ella Friedman Oppenheimer.
Robert Oppenheimer was a brilliant student. He attended the Ethical Culture School in New York City and graduated from the Ethical Culture Fieldston School in 1921. He then attended Harvard University, where he received his bachelor's degree.
[USER]:
I have two apples, each costing 5 dollars. What is the total cost of these apple
[ASSISTANT]:
The total cost of the two apples is 10 dollars.
[USER]:
How about four apples?
[ASSISTANT]:
The total cost of four apples is 20 dollars.
```

## Build and run
Expand All @@ -59,91 +60,119 @@ Second, use `cargo` to build the example project.
cargo build --target wasm32-wasi --release
```

The output WASM file is `target/wasm32-wasi/release/llama-chat.wasm`.

We also need to get the model. Here we use the llama-2-13b model.
The output WASM file is `target/wasm32-wasi/release/llama-chat.wasm`. Next, use WasmEdge to load the llama-2-7b model and then ask the model to questions.

```bash
curl -LO https://huggingface.co/wasmedge/llama2/blob/main/llama-2-13b-q5_k_m.gguf
wasmedge --dir .:. --nn-preload default:GGML:AUTO:llama-2-7b-chat-q5_k_m.gguf llama-chat.wasm
```

Next, use WasmEdge to load the llama-2-13b model and then ask the model to questions.
After executing the command, you may need to wait a moment for the input prompt to appear. You can enter your question once you see the `[USER]:` prompt:

```bash
wasmedge --dir .:. \
--nn-preload default:GGML:CPU:llama-2-13b.Q5_K_M.gguf llama-chat.wasm default \
--prompt 'Robert Oppenheimer most important achievement is ' \
--ctx-size 4096
[USER]:
Who is Robert Oppenheimer?
[ASSISTANT]:
Robert Oppenheimer was an American theoretical physicist and director of the Manhattan Project, which developed the atomic bomb during World War II. He is widely regarded as one of the most important physicists of the 20th century and is known for his contributions to the development of quantum mechanics and the theory of the atomic nucleus. Oppenheimer was also a prominent figure in the post-war nuclear weapons debate and was a strong advocate for international cooperation on nuclear weapons control.
```
After executing the command, you may need to wait a moment for the input prompt to appear. You can enter your question once you see the `[USER]:` prompt:
## Options
You can configure the chat inference application through CLI options.
```bash
Robert Oppenheimer most important achievement is
1945 Manhattan Project.
Robert Oppenheimer was born in New York City on April 22, 1904. He was the son of Julius Oppenheimer, a wealthy German-Jewish textile merchant, and Ella Friedman Oppenheimer.
Robert Oppenheimer was a brilliant student. He attended the Ethical Culture School in New York City and graduated from the Ethical Culture Fieldston School in 1921. He then attended Harvard University, where he received his bachelor's degree
-m, --model-alias <ALIAS>
Model alias [default: default]
-c, --ctx-size <CTX_SIZE>
Size of the prompt context [default: 4096]
-n, --n-predict <N_PRDICT>
Number of tokens to predict [default: 1024]
-g, --n-gpu-layers <N_GPU_LAYERS>
Number of layers to run on the GPU [default: 100]
-b, --batch-size <BATCH_SIZE>
Batch size for prompt processing [default: 4096]
-r, --reverse-prompt <REVERSE_PROMPT>
Halt generation at PROMPT, return control.
-s, --system-prompt <SYSTEM_PROMPT>
System prompt message string [default: "[Default system message for the prompt template]"]
-p, --prompt-template <TEMPLATE>
Prompt template. [default: llama-2-chat] [possible values: llama-2-chat, codellama-instruct, mistral-instruct-v0.1, mistrallite, openchat, belle-llama-2-chat, vicuna-chat, chatml]
--log-prompts
Print prompt strings to stdout
--log-stat
Print statistics to stdout
--log-all
Print all log information to stdout
--stream-stdout
Print the output to stdout in the streaming way
-h, --help
Print help
```
## Optional: Configure the model
The `--prompt-template` option is perhaps the most interesting. It allows the application to support different open source LLM models beyond llama2.
You can use environment variables to configure the model execution.
| Template name | Model | Download |
| ------------ | ------------------------------ | --- |
| llama-2-chat | [The standard llama2 chat model](https://ai.meta.com/llama/) | [7b](https://huggingface.co/wasmedge/llama2/resolve/main/llama-2-7b-chat-q5_k_m.gguf) |
| codellama-instruct | [CodeLlama](https://about.fb.com/news/2023/08/code-llama-ai-for-coding/) | [7b](https://huggingface.co/TheBloke/CodeLlama-7B-Instruct-GGUF/resolve/main/codellama-7b-instruct.Q5_K_M.gguf) |
| mistral-instruct-v0.1 | [Mistral](https://mistral.ai/) | [7b](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q5_K_M.gguf) |
| mistrallite | [Mistral Lite](https://huggingface.co/amazon/MistralLite) | [7b](https://huggingface.co/TheBloke/MistralLite-7B-GGUF/resolve/main/mistrallite.Q5_K_M.gguf) |
| openchat | [OpenChat](https://github.com/imoneoi/openchat) | [7b](https://huggingface.co/TheBloke/openchat_3.5-GGUF/resolve/main/openchat_3.5.Q5_K_M.gguf) |
| belle-llama-2-chat | [BELLE](https://github.com/LianjiaTech/BELLE) | [13b](https://huggingface.co/second-state/BELLE-Llama2-13B-Chat-0.4M-GGUF/resolve/main/BELLE-Llama2-13B-Chat-0.4M-ggml-model-q4_0.gguf) |
| vicuna-chat | [Vicuna](https://lmsys.org/blog/2023-03-30-vicuna/) | [7b](https://huggingface.co/TheBloke/vicuna-7B-v1.5-GGUF/resolve/main/vicuna-7b-v1.5.Q5_K_M.gguf) |
| chatml | [ChatML](https://huggingface.co/chargoddard/rpguild-chatml-13b) | [13b](https://huggingface.co/TheBloke/rpguild-chatml-13B-GGUF/resolve/main/rpguild-chatml-13b.Q5_K_M.gguf) |
| Option |Default |Function |
| -------|-----------|----- |
| LLAMA_LOG | 0 |The backend will print diagnostic information when this value is set to 1|
|LLAMA_N_CTX |512| The context length is the max number of tokens in the entire conversation|
|LLAMA_N_PREDICT |512|The number of tokens to generate in each response from the model|
For example, the following command specifies a context length of 4k tokens, which is standard for llama2, and the max number of tokens in each response to be 1k. It also tells WasmEdge to print out logs and statistics of the model at runtime.
Furthermore, the following command tells WasmEdge to print out logs and statistics of the model at runtime.
```
LLAMA_LOG=1 LLAMA_N_CTX=4096 LLAMA_N_PREDICT=128 wasmedge --dir .:. \
--nn-preload default:GGML:CPU:llama-2-7b.Q5_K_M.gguf llama-simple.wasm default \
--prompt 'Robert Oppenheimer most important achievement is ' \
--ctx-size 4096
...................................................................................................
[2023-10-08 23:13:10.272] [info] [WASI-NN] GGML backend: set n_ctx to 4096
llama_new_context_with_model: kv self size = 2048.00 MB
llama_new_context_with_model: compute buffer total size = 297.47 MB
llama_new_context_with_model: max tensor size = 102.54 MB
[2023-10-08 23:13:10.472] [info] [WASI-NN] GGML backend: llama_system_info: AVX = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 0 | NEON = 1 | ARM_FMA = 1 | F16C = 0 | FP16_VA = 1 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 0 | SSSE3 = 0 | VSX = 0 |
[2023-10-08 23:13:10.472] [info] [WASI-NN] GGML backend: set n_predict to 128
[2023-10-08 23:13:16.014] [info] [WASI-NN] GGML backend: llama_get_kv_cache_token_count 128
llama_print_timings: load time = 1431.58 ms
llama_print_timings: sample time = 3.53 ms / 118 runs ( 0.03 ms per token, 33446.71 tokens per second)
llama_print_timings: prompt eval time = 1230.69 ms / 11 tokens ( 111.88 ms per token, 8.94 tokens per second)
llama_print_timings: eval time = 4295.81 ms / 117 runs ( 36.72 ms per token, 27.24 tokens per second)
llama_print_timings: total time = 5742.71 ms
Robert Oppenheimer most important achievement is
1945 Manhattan Project.
Robert Oppenheimer was born in New York City on April 22, 1904. He was the son of Julius Oppenheimer, a wealthy German-Jewish textile merchant, and Ella Friedman Oppenheimer.
Robert Oppenheimer was a brilliant student. He attended the Ethical Culture School in New York City and graduated from the Ethical Culture Fieldston School in 1921. He then attended Harvard University, where he received his bachelor's degree.
wasmedge --dir .:. --nn-preload default:GGML:AUTO:llama-2-7b-chat-q5_k_m.gguf \
llama-chat.wasm --prompt-template llama-2-chat --log-stat
..................................................................................................
llama_new_context_with_model: n_ctx = 512
llama_new_context_with_model: freq_base = 10000.0
llama_new_context_with_model: freq_scale = 1
llama_new_context_with_model: kv self size = 256.00 MB
llama_new_context_with_model: compute buffer total size = 76.63 MB
[2023-11-07 02:07:44.019] [info] [WASI-NN] GGML backend: llama_system_info: AVX = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 0 | NEON = 1 | ARM_FMA = 1 | F16C = 0 | FP16_VA = 1 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 0 | SSSE3 = 0 | VSX = 0 |

llama_print_timings: load time = 11523.19 ms
llama_print_timings: sample time = 2.62 ms / 102 runs ( 0.03 ms per token, 38961.04 tokens per second)
llama_print_timings: prompt eval time = 11479.27 ms / 49 tokens ( 234.27 ms per token, 4.27 tokens per second)
llama_print_timings: eval time = 13571.37 ms / 101 runs ( 134.37 ms per token, 7.44 tokens per second)
llama_print_timings: total time = 25104.57 ms
[ASSISTANT]:
Ah, a fellow Peanuts enthusiast! Snoopy is Charlie Brown's lovable and imaginative beagle, known for his wild and wacky adventures in the comic strip and television specials. He's a loyal companion to Charlie Brown and the rest of the Peanuts gang, and his antics often provide comic relief in the series. Is there anything else you'd like to know about Snoopy? 🐶
```
## Improve performance
## Improving performance
You can make the inference program run faster by AOT compiling the wasm file first.
```bash
wasmedge compile llama-chat.wasm llama-chat.wasm
wasmedge --dir .:. \
--nn-preload default:GGML:CPU:llama-2-13b-q5_k_m.gguf \
llama-chat.wasm --model-alias default --prompt-template llama-2-chat
wasmedge --dir .:. --nn-preload default:GGML:AUTO:llama-2-7b-chat-q5_k_m.gguf llama-chat.wasm
```
## Understand the code
The [main.rs](https://github.com/second-state/llama-utils/blob/main/chat/src/main.rs
) is the full Rust code to create an interactive chatbot using a LLM. The Rust program manages the user input, tracks the conversation history, transforms the text into the llama2 and other model’s chat templates, and runs the inference operations using the WASI NN standard API.
The [main.rs](https://github.com/second-state/llama-utils/blob/main/chat/src/main.rs) is the full Rust code to create an interactive chatbot using a LLM. The Rust program manages the user input, tracks the conversation history, transforms the text into the llama2 and other model’s chat templates, and runs the inference operations using the WASI NN standard API. The code logic for the chat interaction is somewhat complex. In this section, we will use the [simple example](https://github.com/second-state/llama-utils/tree/main/simple) to explain how to set up and perform one inference round trip. Here is how you use the simple example.
```bash
# Download the compiled simple inference wasm
curl -LO https://github.com/second-state/llama-utils/raw/main/simple/llama-simple.wasm
# Give it a prompt and ask it to use the model to complete it.
wasmedge --dir .:. --nn-preload default:GGML:AUTO:llama-2-7b-chat-q5_k_m.gguf llama-simple.wasm \
--prompt 'Robert Oppenheimer most important achievement is ' --ctx-size 4096
output: in 1942, when he led the team that developed the first atomic bomb, which was dropped on Hiroshima, Japan in 1945.
```
First, let's parse command line arguments to customize the chatbot's behavior using `Command` struct. It extracts the following parameters: `prompt` (a prompt that guides the conversation), `model_alias` (a list for the loaded model), and `ctx_size` (the size of the chat context).
```rust
fn main() -> Result<(), String> {
let matches = Command::new("Llama API Server")
let matches = Command::new("Simple LLM inference")
.arg(
Arg::new("prompt")
.short('p')
Expand Down Expand Up @@ -216,14 +245,14 @@ Next, The prompt is converted into bytes and set as the input tensor for the mod
.expect("Failed to set prompt as the input tensor");
```
Next, excute the model inference.
Next, execute the model inference.
```rust
// execute the inference
context.compute().expect("Failed to complete inference");
```
After the inference is fiished, extract the result from the computation context and losing invalid UTF8 sequences handled by converting the output to a string using `String::from_utf8_lossy`.
After the inference is finished, extract the result from the computation context and losing invalid UTF8 sequences handled by converting the output to a string using `String::from_utf8_lossy`.
```rust
let mut output_buffer = vec![0u8; *CTX_SIZE.get().unwrap()];
Expand All @@ -241,8 +270,9 @@ println!("\nprompt: {}", &prompt);
println!("\noutput: {}", output);
```
The code explanation above is simple one time chat with llama 2 model. But we have more!
## Resources
* If you're looking for multi-turn conversations with llama 2 models, please check out the above mentioned chat example source code [here](https://github.com/second-state/llama-utils/tree/main/chat).
* If you want to construct OpenAI-compatible APIs specifically for your llama2 model, or the Llama2 model itself, please check out the source code [for the API server](https://github.com/second-state/llama-utils/tree/main/api-server).
* To learn more, please check out [this article](https://medium.com/stackademic/fast-and-portable-llama2-inference-on-the-heterogeneous-edge-a62508e82359).

* If you're looking for continuous conversations with llama 2 models, please check out the source code [here](https://github.com/second-state/llama-utils/tree/main/chat).
* If you want to construct OpenAI-compatible APIs specifically for your llama2 model, or the Llama2 model itself, please check out the surce code [here](https://github.com/second-state/llama-utils/tree/main/api-server).
* For the reason why we need to run LLama2 model with WasmEdge, please check out [this article](https://medium.com/stackademic/fast-and-portable-llama2-inference-on-the-heterogeneous-edge-a62508e82359).
Loading

0 comments on commit 514afc1

Please sign in to comment.