added missing openai values in settings (#13)

3x3cut0r · May 22, 2024 · 5e8e0fa · 5e8e0fa
1 parent 0c20e9b
commit 5e8e0fa
Show file tree

Hide file tree

Showing 4 changed files with 57 additions and 27 deletions.
diff --git a/privategpt/Dockerfile b/privategpt/Dockerfile
@@ -142,10 +142,15 @@ ENV PYTHONPATH="$PYTHONPATH:/private_gpt/" \
     OPENAI_API_BASE="https://api.openai.com/v1" \
     OPENAI_API_KEY="sk-1234" \
     OPENAI_MODEL="gpt-3.5-turbo" \
-    OLLAMA_LLM_MODEL="mistral:latest" \
-    OLLAMA_EMBEDDING_MODEL="" \
+    OPENAI_REQUEST_TIMEOUT="120.0" \
+    OPENAI_EMBEDDING_API_BASE="" \
+    OPENAI_EMBEDDING_API_KEY="" \ 
+    OPENAI_EMBEDDING_MODEL="text-embedding-3-small" \
     OLLAMA_API_BASE="http://localhost:11434" \
-    OLLAMA_EMBEDDING_API_BASE="http://localhost:11434" \
+    OLLAMA_EMBEDDING_API_BASE="" \
+    OLLAMA_LLM_MODEL="mistral:latest" \
+    OLLAMA_EMBEDDING_MODEL="nomic-embed-text" \
+    OLLAMA_KEEP_ALIVE="5m" \
     OLLAMA_TFS_Z="1.0" \
     OLLAMA_NUM_PREDICT="128" \
     OLLAMA_TOP_K="40" \

diff --git a/privategpt/README.md b/privategpt/README.md
@@ -220,13 +220,18 @@ secret: "Basic c2VjcmV0OmtleQ=="
 - `OPENAI_API_BASE` - Base URL of OpenAI API. Example: https://api.openai.com/v1 - **Default: https://api.openai.com/v1**
 - `OPENAI_API_KEY` - Your API Key for the OpenAI API. Example: sk-1234 - **Default: sk-1234**
 - `OPENAI_MODEL` - OpenAI Model to use. (see [OpenAI Models Overview](https://platform.openai.com/docs/models/overview)). Example: gpt-4 - **Default: gpt-3.5-turbo**
+- `OPENAI_REQUEST_TIMEOUT` - Time elapsed until openailike server times out the request. Default is 120s. Format is float. - **Default: 120.0**
+- `OPENAI_EMBEDDING_API_BASE` - Base URL of OpenAI API. Example: https://api.openai.com/v1 - **Default: same as OPENAI_API_BASE**
+- `OPENAI_EMBEDDING_API_KEY` - Your API Key for the OpenAI Embedding API. Example: sk-1234. - **Default: same as OPENAI_API_KEY**
+- `OPENAI_EMBEDDING_MODEL` - OpenAI embedding Model to use. Example: text-embedding-3-large - **Default: text-embedding-3-small**
 
 ###### Ollama
 
+- `OLLAMA_API_BASE` - Base URL of Ollama API. Example: http://192.168.1.100:11434 - **Default: http://localhost:11434**
+- `OLLAMA_EMBEDDING_API_BASE` - Base URL of Ollama Embedding API. Example: http://192.168.1.100:11434 - **Default: same as OLLAMA_API_BASE**
 - `OLLAMA_LLM_MODEL` - Ollama model to use. (see [Ollama Library](https://ollama.com/library)). Example: 'llama2-uncensored' - **Default: mistral:latest**
 - `OLLAMA_EMBEDDING_MODEL` - Model to use. Example: 'nomic-embed-text'. - **Default: nomic-embed-text**
-- `OLLAMA_API_BASE` - Base URL of Ollama API. Example: http://192.168.1.100:11434 - **Default: http://localhost:11434**
-- `OLLAMA_EMBEDDING_API_BASE` - Base URL of Ollama Embedding API. Example: http://192.168.1.100:11434 - **Default: http://localhost:11434**
+- `OLLAMA_KEEP_ALIVE` - Time the model will stay loaded in memory after a request. examples: 5m, 5h, '-1' - **Default: 5m**
 - `OLLAMA_TFS_Z` - Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting. - **Default: 1.0**
 - `OLLAMA_NUM_PREDICT` - Maximum number of tokens to predict when generating text. (Default: 128, -1 = infinite generation, -2 = fill context) - **Default: 128**
 - `OLLAMA_TOP_K` - Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative. - **Default: 40**

diff --git a/privategpt/docker-entrypoint.sh b/privategpt/docker-entrypoint.sh
@@ -186,59 +186,74 @@ sed -i "99s|^.*$|  api_key: ${OPENAI_API_KEY:-"sk-1234"}|" /home/worker/app/sett
 # OPENAI_MODEL
 sed -i "100s|^.*$|  model: ${OPENAI_MODEL:-"gpt-3.5-turbo"}|" /home/worker/app/settings.yaml
 
-# OLLAMA_LLM_MODEL
-sed -i "103s|^.*$|  llm_model: ${OLLAMA_LLM_MODEL:-"mistral:latest"}|" /home/worker/app/settings.yaml
+# OPENAI_REQUEST_TIMEOUT
+sed -i "101s|^.*$|  request_timeout: ${OPENAI_REQUEST_TIMEOUT:-"120.0"}|" /home/worker/app/settings.yaml
 
-# OLLAMA_EMBEDDING_MODEL
-sed -i "104s|^.*$|  embedding_model: ${OLLAMA_EMBEDDING_MODEL:-"nomic-embed-text"}|" /home/worker/app/settings.yaml
+# OPENAI_EMBEDDING_API_BASE
+sed -i "102s|^.*$|  embedding_api_base: ${OPENAI_EMBEDDING_API_BASE:-$OPENAI_API_BASE}|" /home/worker/app/settings.yaml
+
+# OPENAI_EMBEDDING_API_KEY
+sed -i "103s|^.*$|  embedding_api_key: ${OPENAI_EMBEDDING_API_KEY:-$OPENAI_API_KEY}|" /home/worker/app/settings.yaml
+
+# OPENAI_EMBEDDING_MODEL
+sed -i "104s|^.*$|  embedding_model: ${OPENAI_EMBEDDING_MODEL:-"text-embedding-3-small"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_API_BASE
-sed -i "105s|^.*$|  api_base: ${OLLAMA_API_BASE:-"http://localhost:11434"}|" /home/worker/app/settings.yaml
+sed -i "107s|^.*$|  api_base: ${OLLAMA_API_BASE:-"http://localhost:11434"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_EMBEDDING_API_BASE
-sed -i "106s|^.*$|  embedding_api_base: ${OLLAMA_EMBEDDING_API_BASE:-"http://localhost:11434"}|" /home/worker/app/settings.yaml
+sed -i "108s|^.*$|  embedding_api_base: ${OLLAMA_EMBEDDING_API_BASE:-$OLLAMA_API_BASE}|" /home/worker/app/settings.yaml
+
+# OLLAMA_LLM_MODEL
+sed -i "109s|^.*$|  llm_model: ${OLLAMA_LLM_MODEL:-"mistral:latest"}|" /home/worker/app/settings.yaml
+
+# OLLAMA_EMBEDDING_MODEL
+sed -i "110s|^.*$|  embedding_model: ${OLLAMA_EMBEDDING_MODEL:-"nomic-embed-text"}|" /home/worker/app/settings.yaml
+
+# OLLAMA_KEEP_ALIVE
+sed -i "111s|^.*$|  embedding_model: ${OLLAMA_KEEP_ALIVE:-"5m"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_TFS_Z
-sed -i "107s|^.*$|  tfs_z: ${OLLAMA_TFS_Z:-"1.0"}|" /home/worker/app/settings.yaml
+sed -i "112s|^.*$|  tfs_z: ${OLLAMA_TFS_Z:-"1.0"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_NUM_PREDICT
-sed -i "108s|^.*$|  num_predict: ${OLLAMA_NUM_PREDICT:-"128"}|" /home/worker/app/settings.yaml
+sed -i "113s|^.*$|  num_predict: ${OLLAMA_NUM_PREDICT:-"128"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_TOP_K
-sed -i "109s|^.*$|  top_k: ${OLLAMA_TOP_K:-"40"}|" /home/worker/app/settings.yaml
+sed -i "114s|^.*$|  top_k: ${OLLAMA_TOP_K:-"40"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_TOP_P
-sed -i "110s|^.*$|  top_p: ${OLLAMA_TOP_P:-"0.9"}|" /home/worker/app/settings.yaml
+sed -i "115s|^.*$|  top_p: ${OLLAMA_TOP_P:-"0.9"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_REPEAT_LAST_N
-sed -i "111s|^.*$|  repeat_last_n: ${OLLAMA_REPEAT_LAST_N:-"64"}|" /home/worker/app/settings.yaml
+sed -i "116s|^.*$|  repeat_last_n: ${OLLAMA_REPEAT_LAST_N:-"64"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_REPEAT_PENALTY
-sed -i "112s|^.*$|  repeat_penalty: ${OLLAMA_REPEAT_PENALTY:-"1.1"}|" /home/worker/app/settings.yaml
+sed -i "117s|^.*$|  repeat_penalty: ${OLLAMA_REPEAT_PENALTY:-"1.1"}|" /home/worker/app/settings.yaml
 
 # OLLAMA_REQUEST_TIMEOUT
-sed -i "113s|^.*$|  request_timeout: ${OLLAMA_REQUEST_TIMEOUT:-"120.0"}|" /home/worker/app/settings.yaml
+sed -i "118s|^.*$|  request_timeout: ${OLLAMA_REQUEST_TIMEOUT:-"120.0"}|" /home/worker/app/settings.yaml
 
 # AZOPENAI_API_KEY
-sed -i "116s|^.*$|  api_key: ${AZOPENAI_API_KEY:-"sk-1234"}|" /home/worker/app/settings.yaml
+sed -i "121s|^.*$|  api_key: ${AZOPENAI_API_KEY:-"sk-1234"}|" /home/worker/app/settings.yaml
 
 # AZOPENAI_AZURE_ENDPOINT
-sed -i "117s|^.*$|  azure_endpoint: ${AZOPENAI_AZURE_ENDPOINT:-"https://api.myazure.com/v1"}|" /home/worker/app/settings.yaml
+sed -i "122s|^.*$|  azure_endpoint: ${AZOPENAI_AZURE_ENDPOINT:-"https://api.myazure.com/v1"}|" /home/worker/app/settings.yaml
 
 # AZOPENAI_API_VERSION
-sed -i "118s|^.*$|  api_version: \"${AZOPENAI_API_VERSION:-"2023_05_15"}\"|" /home/worker/app/settings.yaml
+sed -i "123s|^.*$|  api_version: \"${AZOPENAI_API_VERSION:-"2023_05_15"}\"|" /home/worker/app/settings.yaml
 
 # AZOPENAI_EMBEDDING_DEPLOYMENT_NAME
-sed -i "119s|^.*$|  embedding_deployment_name: ${AZOPENAI_EMBEDDING_DEPLOYMENT_NAME:-"my-azure-embedding-deployment-name"}|" /home/worker/app/settings.yaml
+sed -i "124s|^.*$|  embedding_deployment_name: ${AZOPENAI_EMBEDDING_DEPLOYMENT_NAME:-"my-azure-embedding-deployment-name"}|" /home/worker/app/settings.yaml
 
 # AZOPENAI_EMBEDDING_MODEL
-sed -i "120s|^.*$|  embedding_model: ${AZOPENAI_EMBEDDING_MODEL:-"text-embedding-3-small"}|" /home/worker/app/settings.yaml
+sed -i "125s|^.*$|  embedding_model: ${AZOPENAI_EMBEDDING_MODEL:-"text-embedding-3-small"}|" /home/worker/app/settings.yaml
 
 # AZOPENAI_LLM_DEPLOYMENT_NAME
-sed -i "121s|^.*$|  llm_deployment_name: ${AZOPENAI_LLM_DEPLOYMENT_NAME:-"my-azure-llm-deployment-name"}|" /home/worker/app/settings.yaml
+sed -i "126s|^.*$|  llm_deployment_name: ${AZOPENAI_LLM_DEPLOYMENT_NAME:-"my-azure-llm-deployment-name"}|" /home/worker/app/settings.yaml
 
 # AZOPENAI_LLM_MODEL
-sed -i "122s|^.*$|  llm_model: ${AZOPENAI_LLM_MODEL:-"gpt-4"}|" /home/worker/app/settings.yaml
+sed -i "127s|^.*$|  llm_model: ${AZOPENAI_LLM_MODEL:-"gpt-4"}|" /home/worker/app/settings.yaml
 
 ############################
 # run app                  #

diff --git a/privategpt/settings_template.yaml b/privategpt/settings_template.yaml
@@ -98,12 +98,17 @@ openai:
   api_base: https://api.openai.com/v1
   api_key: sk-1234
   model: gpt-3.5-turbo
+  request_timeout: 120.0
+  embedding_api_base: https://api.openai.com/v1
+  embedding_api_key: sk-1234
+  embedding_model: text-embedding-3-small
 
 ollama:
-  llm_model: mistral:latest
-  embedding_model: nomic-embed-text
   api_base: http://localhost:11434
   embedding_api_base: http://localhost:11434
+  llm_model: mistral:latest
+  embedding_model: nomic-embed-text
+  keep_alive: 5m
   tfs_z: 1.0
   num_predict: 128
   top_k: 40