From 2a4cd8fdfdb59ba745de3b8156661817014e29c4 Mon Sep 17 00:00:00 2001 From: Artem Zahumonnyi Date: Fri, 28 Aug 2026 09:34:51 +0300 Subject: [PATCH] docs(config): remove deprecated o3 and o3-mini model references --- .../ai-models-integration/azure-openai.md | 4 +- .../codemie-native-llm-config.md | 43 +----------------- .../codemie/api-configuration.md | 2 +- .../litellm-proxy/model-configuration.md | 44 ------------------- faq/list-of-models.md | 2 +- 5 files changed, 5 insertions(+), 90 deletions(-) diff --git a/docs/admin/configuration/codemie/ai-models-integration/azure-openai.md b/docs/admin/configuration/codemie/ai-models-integration/azure-openai.md index dd258c93..a65b161a 100644 --- a/docs/admin/configuration/codemie/ai-models-integration/azure-openai.md +++ b/docs/admin/configuration/codemie/ai-models-integration/azure-openai.md @@ -11,7 +11,7 @@ pagination_next: null ## Overview -This guide explains how to enable Azure OpenAI services in your Azure account and deploy AI models for use with AI/Run CodeMie. Azure OpenAI provides access to OpenAI models including GPT-4.1, GPT-5, o1, and o3-series models. +This guide explains how to enable Azure OpenAI services in your Azure account and deploy AI models for use with AI/Run CodeMie. Azure OpenAI provides access to OpenAI models including GPT-4.1, GPT-5, and o1. :::info When to Use This Guide This configuration is required if you plan to use Azure OpenAI models such as GPT-4.1, GPT-5, or any OpenAI models hosted on Azure. @@ -139,7 +139,7 @@ Repeat **Step 3** to deploy additional models based on your requirements: - **GPT-4o-mini**: Cost-effective option for simpler tasks - **GPT-4.1**: Advanced reasoning with larger context window - **GPT-5**: Latest generation model with enhanced capabilities -- **o1/o3-mini**: Specialized reasoning models for complex problem-solving +- **o1**: Specialized reasoning model for complex problem-solving - **Embedding models**: `text-embedding-ada-002` for vector embeddings Each model deployment allows you to set specific TPM limits and deployment configurations. diff --git a/docs/admin/configuration/codemie/ai-models-integration/codemie-native-llm-config.md b/docs/admin/configuration/codemie/ai-models-integration/codemie-native-llm-config.md index 38bdbdd1..80f41f03 100644 --- a/docs/admin/configuration/codemie/ai-models-integration/codemie-native-llm-config.md +++ b/docs/admin/configuration/codemie/ai-models-integration/codemie-native-llm-config.md @@ -212,26 +212,6 @@ llm_models: output: 0.0000004 cache_read_input_token_cost: 0.000000005 - - base_name: "o3-mini" - deployment_name: "o3-mini-2025-01-31" - label: "o3 Mini 2025-01-31" - multimodal: false - react_agent: false - enabled: true - provider: "azure_openai" - max_output_tokens: 100000 - features: - streaming: false - tools: true - temperature: false - parallel_tool_calls: false - system_prompt: false - max_tokens: false - cost: - input: 0.0000011 - output: 0.0000044 - cache_read_input_token_cost: 0.00000055 - - base_name: "o1" deployment_name: "o1-2024-12-17" label: "o1 2024-12-17" @@ -252,27 +232,6 @@ llm_models: output: 0.00006 cache_read_input_token_cost: 0.0000075 - - base_name: "o3-2025-04-16" - deployment_name: "o3-2025-04-16" - label: "o3 2025-04-16" - multimodal: true - react_agent: false - enabled: true - provider: "azure_openai" - max_output_tokens: 100000 - features: - streaming: true - tools: true - temperature: false - parallel_tool_calls: false - system_prompt: false - max_tokens: false - reasoning: true - cost: - input: 0.000002 - output: 0.000008 - cache_read_input_token_cost: 0.0000005 - - base_name: "o4-mini-2025-04-16" deployment_name: "o4-mini-2025-04-16" label: "o4-mini 2025-04-16" @@ -312,7 +271,7 @@ embeddings_models: Standard models (`gpt-4.1`, Claude, Gemini) support parallel tool calls — the agent can issue multiple tool calls in one inference round and stream their results concurrently. -Reasoning models (`o1`, `o3`, `o3-mini`, `o4-mini`, and similar) do **not** support the +Reasoning models (`o1`, `o4-mini`, and similar) do **not** support the `parallel_tool_calls` OpenAI parameter. Sending it causes an API error. Always set `parallel_tool_calls: false` in the `features` block for these models, as shown in the Azure examples above. When this flag is `false`, the platform automatically strips the diff --git a/docs/admin/configuration/codemie/api-configuration.md b/docs/admin/configuration/codemie/api-configuration.md index fdaca620..f809b9d3 100644 --- a/docs/admin/configuration/codemie/api-configuration.md +++ b/docs/admin/configuration/codemie/api-configuration.md @@ -1282,7 +1282,7 @@ tool calls simultaneously within one inference round. Results arrive concurrentl rendered in the UI as parallel entries under the same thought step. Standard GPT and Claude models support parallel tool calls. Reasoning models -(`o1`, `o3`, `o3-mini`, `o4-mini`, and similar) do **not** — always set +(`o1`, `o4-mini`, and similar) do **not** — always set `parallel_tool_calls: false` in their `features` block. ::: diff --git a/docs/admin/configuration/extensions/litellm-proxy/model-configuration.md b/docs/admin/configuration/extensions/litellm-proxy/model-configuration.md index 8b62c8f1..360a13af 100644 --- a/docs/admin/configuration/extensions/litellm-proxy/model-configuration.md +++ b/docs/admin/configuration/extensions/litellm-proxy/model-configuration.md @@ -342,8 +342,6 @@ Configuration examples for these models can be found in the provider-specific se | [`gpt-5.4-2026-03-05`](#gpt-54) | GPT-5.4 | | [`gpt-5.5-2026-04-24`](#gpt-55) | GPT-5.5 | | [`o1`](#o1) | o1 | -| [`o3-mini`](#o3-mini) | o3 mini | -| [`o3-2025-04-16`](#o3) | o3 | | [`o4-mini-2025-04-16`](#o4-mini) | o4 mini | | [`codemie-text-embedding-ada-002`](#text-embedding-ada-002) | Text Embedding Ada-002 | | [`codemie-text-embedding-3-small`](#text-embedding-3-small) | Text Embedding 3 Small | @@ -1169,48 +1167,6 @@ model_list: -#### o3 - -
-o3 - -```yaml -- model_name: o3-2025-04-16 - litellm_params: - model: azure/o3-2025-04-16 - api_base: https://api-base-swedencentral-0.openai.azure.com/ - litellm_credential_name: default_azure_openai_credential - api_version: 2024-12-01-preview - model_info: - id: o3-2025-04-16-swedencentral-0-eu - base_model: azure/o3-2025-04-16 - label: "o3 2025-04-16" - supports_native_streaming: false -``` - -
- -#### o3-Mini - -
-o3-Mini - -```yaml -- model_name: o3-mini - litellm_params: - model: azure/o3-mini-2025-01-31 - api_base: https://api-base-swedencentral-0.openai.azure.com/ - litellm_credential_name: default_azure_openai_credential - api_version: 2024-12-01-preview - model_info: - id: o3-mini-swedencentral-0-eu - base_model: azure/eu/o3-mini-2025-01-31 - label: "o3 Mini 2025-01-31" - supports_native_streaming: false -``` - -
- #### o4-Mini
diff --git a/faq/list-of-models.md b/faq/list-of-models.md index acbd2281..2047ecd4 100644 --- a/faq/list-of-models.md +++ b/faq/list-of-models.md @@ -1,6 +1,6 @@ # List of models? What AI models are available in CodeMie? LLM model list? -GPT-4o, GPT-4o 2024-08-06, GPT-4o 2024-11-20, GPT-4o-mini-2024-07-18, Gemini 2.0 Flash Exp, Bedrock Claude 3.5 Sonnet, Bedrock Claude 3.5 Sonnet v2 20241022, Bedrock Claude 3.7, o3 Mini 2025-01-31, o3 Mini 2025-01-31, Mixtral 8x7B Instruct +GPT-4o, GPT-4o 2024-08-06, GPT-4o 2024-11-20, GPT-4o-mini-2024-07-18, Gemini 2.0 Flash Exp, Bedrock Claude 3.5 Sonnet, Bedrock Claude 3.5 Sonnet v2 20241022, Bedrock Claude 3.7, Mixtral 8x7B Instruct Note: The list of models is updated very often, so it will be easier for you to check it on the UI