Add MiniMax as a supported LLM provider

MiniMax offers OpenAI-compatible API with models like MiniMax-M2.5 and
MiniMax-M2.5-highspeed (204K context window). This commit adds:

- MiniMax provider section in README (setup guide, cost estimation)
- Example config file (configs/minimax_config.yaml)
- Updated test env to support MiniMax API key validation

Co-Authored-By: octo-patch <octo-patch@users.noreply.github.com>

Author: octo-patch

README.md

@@ -202,7 +202,7 @@ OpenEvolve implements a sophisticated **evolutionary coding pipeline** that goes
 <details>
 <summary><b>Advanced LLM Integration</b></summary>
 
-- **Universal API**: Works with OpenAI, Google, local models, and proxies
+- **Universal API**: Works with OpenAI, Google, MiniMax, local models, and proxies
 - **Intelligent Ensembles**: Weighted combinations with sophisticated fallback
 - **Test-Time Compute**: Enhanced reasoning through proxy systems (see [OptiLLM setup](#llm-provider-setup))
 - **Plugin Ecosystem**: Support for advanced reasoning plugins
@@ -281,6 +281,7 @@ docker run --rm -v $(pwd):/app ghcr.io/algorithmicsuperintelligence/openevolve:l
 - **o3-mini**: ~$0.03-0.12 per iteration (more cost-effective)
 - **Gemini-2.5-Pro**: ~$0.08-0.30 per iteration
 - **Gemini-2.5-Flash**: ~$0.01-0.05 per iteration (fastest and cheapest)
+- **MiniMax-M2.5**: ~$0.02-0.08 per iteration (204K context, OpenAI-compatible)
 - **Local models**: Nearly free after setup
 - **OptiLLM**: Use cheaper models with test-time compute for better results
 
@@ -320,6 +321,33 @@ export OPENAI_API_KEY="your-gemini-api-key"
 
 </details>
 
+<details>
+<summary><b>🧠 MiniMax</b></summary>
+
+[MiniMax](https://www.minimaxi.com/) offers powerful models with 204K context window via an OpenAI-compatible API:
+
+```yaml
+# config.yaml
+llm:
+  api_base: "https://api.minimax.io/v1"
+  api_key: "${MINIMAX_API_KEY}"
+  models:
+    - name: "MiniMax-M2.5"
+      weight: 0.6
+    - name: "MiniMax-M2.5-highspeed"
+      weight: 0.4
+```
+
+```bash
+export MINIMAX_API_KEY="your-minimax-api-key"
+```
+
+> **Note:** MiniMax requires temperature to be in (0.0, 1.0] — zero is not accepted. The default 0.7 works well.
+
+See [`configs/minimax_config.yaml`](configs/minimax_config.yaml) for a complete configuration example.
+
+</details>
+
 <details>
 <summary><b>🏠 Local Models (Ollama/vLLM)</b></summary>
 
@@ -792,7 +820,7 @@ See the [Cost Estimation](#cost-estimation) section in Installation & Setup for
 
 **Yes!** OpenEvolve supports any OpenAI-compatible API:
 
-- **Commercial**: OpenAI, Google, Cohere
+- **Commercial**: OpenAI, Google, Cohere, MiniMax
 - **Local**: Ollama, vLLM, LM Studio, text-generation-webui
 - **Advanced**: OptiLLM for routing and test-time compute
 

configs/README.md

@@ -12,6 +12,9 @@ The main configuration file containing all available options with sensible defau
 
 Use this file as a template for your own configurations.
 
+### `minimax_config.yaml`
+A complete configuration for using [MiniMax](https://www.minimaxi.com/) models (MiniMax-M2.5, MiniMax-M2.5-highspeed) with OpenEvolve. MiniMax provides an OpenAI-compatible API with 204K context window support.
+
 ### `island_config_example.yaml`
 A practical example configuration demonstrating proper island-based evolution setup. Shows:
 - Recommended island settings for most use cases

configs/minimax_config.yaml

@@ -0,0 +1,74 @@
+# OpenEvolve Configuration for MiniMax
+# MiniMax provides OpenAI-compatible API with powerful models like MiniMax-M2.5
+# Get your API key from: https://platform.minimaxi.com/
+#
+# Set your API key:
+#   export MINIMAX_API_KEY="your-minimax-api-key"
+
+# General settings
+max_iterations: 100
+checkpoint_interval: 10
+log_level: "INFO"
+random_seed: 42
+
+# LLM configuration for MiniMax
+llm:
+  api_base: "https://api.minimax.io/v1"
+  api_key: "${MINIMAX_API_KEY}"
+
+  # MiniMax models for evolution
+  models:
+    - name: "MiniMax-M2.5"
+      weight: 0.6
+    - name: "MiniMax-M2.5-highspeed"
+      weight: 0.4
+
+  # MiniMax models for LLM feedback
+  evaluator_models:
+    - name: "MiniMax-M2.5-highspeed"
+      weight: 1.0
+
+  # Generation parameters
+  # Note: MiniMax requires temperature to be in (0.0, 1.0] — zero is not accepted
+  temperature: 0.7
+  top_p: 0.95
+  max_tokens: 4096
+
+  # Request parameters
+  timeout: 120
+  retries: 3
+  retry_delay: 5
+
+# Evolution settings
+diff_based_evolution: true
+max_code_length: 10000
+
+# Prompt configuration
+prompt:
+  system_message: "You are an expert coder helping to improve programs through evolution."
+  evaluator_system_message: "You are an expert code reviewer."
+  num_top_programs: 3
+  num_diverse_programs: 2
+  use_template_stochasticity: true
+  include_artifacts: true
+
+# Database configuration
+database:
+  population_size: 1000
+  num_islands: 5
+  migration_interval: 50
+  migration_rate: 0.1
+  feature_dimensions:
+    - "complexity"
+    - "diversity"
+  feature_bins: 10
+
+# Evaluator configuration
+evaluator:
+  timeout: 300
+  cascade_evaluation: true
+  cascade_thresholds:
+    - 0.5
+    - 0.75
+    - 0.9
+  parallel_evaluations: 4

tests/test_valid_configs.py

@@ -24,7 +24,7 @@ def collect_files(self):
                     config_files.append(os.path.join(root, file))
         return config_files
 
-    @patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key-for-validation"})
+    @patch.dict(os.environ, {"ANTHROPIC_API_KEY": "test-key-for-validation", "MINIMAX_API_KEY": "test-key-for-validation"})
     def test_import_config_files(self):
         """Attempt to import all config files"""
         config_files = self.collect_files()