Purpose of the change

OpenAI is deprecating all GPT-4.x models effective February 13th, 2026. This PR replaces all occurrences of gpt-4* with their gpt-5* equivalents.

Description

Here are the model mappings and the files touched. The model names were obtained from https://developers.openai.com/api/docs/models.

This is a search-and-replace PR. No other changes were made.

Model Mapping

Files Changed (34 files)

Source Code (3 files)

• src/memmachine/installation/utilities.py — DEFAULT_OPENAI_MODEL default
• src/memmachine/common/api/config_spec.py — OpenAI Responses and Chat Completions model defaults
• src/memmachine/main/memmachine.py — Short-term memory LLM default

Shell Scripts (1 file)

• memmachine-compose.sh — OpenAI model prompt, default fallback, and help text (4 references)

Sample Configs (2 files)

• sample_configs/episodic_memory_config.gpu.sample
• sample_configs/episodic_memory_config.cpu.sample

Documentation (4 files)

• docs/open_source/configuration.mdx — Config examples and parameter table
• docs/install_guide/install_guide.mdx — Install guide YAML examples
• docs/install_guide/cloud_deploy/aws_cloudformation.mdx — CloudFormation template
• docs/examples/agents.mdx — Agent example env var

Examples (7 files)

• examples/simple_chatbot/app.py — Default model constant
• examples/simple_chatbot/llm.py — Model string default
• examples/simple_chatbot/model_config.py — Provider model map and display names
• examples/simple_chatbot/README.md — Multi-provider LLM list
• examples/v1/frontend/llm.py — Model string default
• examples/v1/frontend/model_config.py — Provider model map
• examples/v1/crm/slack_server.py — OpenAI chat completion model
• examples/v1/crm/README.md — Feature description

Evaluation Scripts (6 files)

• evaluation/locomo/episodic_agent/memmachine_locomo.py
• evaluation/locomo/episodic_agent/restapiv2_locomo_search_agent.py
• evaluation/locomo/episodic_agent/llm_judge.py
• evaluation/locomo/episodic_memory/locomo_search.py
• evaluation/locomo/episodic_memory/restapiv2_locomo_search.py
• evaluation/locomo/episodic_memory/llm_judge.py

Tests (11 files)

• tests/memmachine/conftest.py — Integration config fixture
• tests/memmachine/main/test_memmachine_mock.py — STM default assertion
• tests/memmachine/installation/test_configuration_wizard.py — Wizard prompt test input
• tests/memmachine/common/configuration/test_language_model_conf.py — Model config fixtures and assertions
• tests/memmachine/common/resource_manager/test_language_model_manager.py — Manager mock config
• tests/memmachine/rest_client/test_config.py — Semantic memory config tests
• tests/memmachine/rest_client/test_memory.py — Semantic set configure tests
• tests/memmachine/server/api_v2/test_config_router.py — Resource and memory config endpoint tests
• tests/memmachine/server/api_v2/test_config_service.py — Config service persistence test
• tests/memmachine/server/api_v2/test_router.py — Semantic set configure endpoint test

Fixes/Closes

Fixes #1005

Type of change

• Project Maintenance (updates to build scripts, CI, etc., that do not affect the main project)

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration.

I only verified that MemMachine starts. I HAVE NOT done any end-to-end testing or QA. This needs to be done using the nightly testing before approval and merging.

• Manual verification (list step-by-step instructions)

The following grep found no mentions of GPT 4 models in the source code.

$ grep -Rsi "gpt-4" *

Checklist

• I have signed the commit(s) within this pull request
• I have performed a self-review of my own code
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• New and existing unit tests pass locally with my changes
• I have checked my code and corrected any misspellings

Maintainer Checklist

• Confirmed all checks passed
• Contributor has signed the commit(s)
• Reviewed the code
• Run, Tested, and Verified the change(s) work as expected

Screenshots/Gifs

N/A

Further comments

None