Add Vicuna

api/evaluator_app.py

@@ -17,6 +17,9 @@ import pandas as pd
 from typing import Dict, List
 from json import JSONDecodeError
 from langchain.llms import Anthropic
+from langchain.llms import LlamaCpp
+from langchain.callbacks.base import CallbackManager
+from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 from langchain.chat_models import ChatAnthropic
 from langchain.schema import BaseRetriever, Document
 from langchain.chains.question_answering import load_qa_chain
@@ -103,6 +106,13 @@ def make_llm(model):
         llm = Anthropic(temperature=0)
     elif model == "Anthropic-100k":
         llm = Anthropic(model="claude-v1-100k",temperature=0)
+        llm = ChatAnthropic(temperature=0)
+    elif model in ("vicuna-7b","vicuna-13b"):
+        callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
+        if model == "vicuna-7b":
+            llm = LlamaCpp(model_path="/Users/31treehaus/Desktop/AI/llama.cpp/models/vicuna_7B/ggml-vicuna-7b-q4_0.bin", callback_manager=callback_manager, n_threads=16, n_ctx=10000, verbose=True, temperature=0)
+        else:
+            llm = LlamaCpp(model_path="/Users/31treehaus/Desktop/AI/llama.cpp/models/vicuna_13B/ggml-vicuna-13b-4bit.bin", callback_manager=callback_manager, n_threads=16, n_ctx=10000, verbose=True, temperature=0)
     return llm
 
 

api/requirements.txt

@@ -8,7 +8,6 @@ tiktoken==0.3.1
 faiss-cpu==1.7.3
 huggingface-hub==0.12.0
 anthropic==0.2.8
-llama-cpp-python==0.1.32
 pypdf==3.7.1
 filetype==1.2.0
 tokenizers==0.13.3
@@ -19,4 +18,5 @@ sse_starlette==1.3.3
 gpt-index==0.5.16
 faiss-cpu==1.7.3
 python-dotenv==1.0.0
-sentry_sdk==1.21.1
+sentry_sdk==1.21.1
+llama-cpp-python==0.1.43

nextjs/components/Playground.tsx

@@ -302,7 +302,7 @@ const Playground = ({ form }: { form: Form }) => {
         title="Instructions"
         style={alertStyle}
       >
-        Upload a text file (up to 50 MB) and choose the parameters for your QA
+        Upload up to 10 text files (up to 50 MB each) and choose the parameters for your QA
         chain. This evaluator will generate a test dataset of QA pairs and grade
         the performance of the QA chain. You can experiment with different
         parameters and evaluate the performance.
@@ -319,8 +319,8 @@ const Playground = ({ form }: { form: Form }) => {
             setTestFilesDropzoneDisabled(false);
             setFileUploadDisabled(true);
           }}
-          maxFiles={1}
-          multiple={false}
+          maxFiles={10}
+          multiple={true}
           maxSize={MAX_FILE_SIZE_MB * 1024 ** 2} // 50 MB
           accept={[
             MIME_TYPES.pdf,

nextjs/components/Sidebar.tsx

@@ -98,6 +98,8 @@ const Sidebar = ({ form }: { form: Form }) => {
                     { label: "GPT 3.5 Turbo", value: "gpt-3.5-turbo" },
                     { label: "GPT 4", value: "gpt-4" },
                     { label: "Anthropic", value: "anthropic" },
+                    { label: "Vicuna 7B", value: "vicuna-7b" },
+                    { label: "Vicuna 13B", value: "vicuna-13b" },
                   ]}
                 />
               )}