diff --git a/README.md b/README.md
index ab4c75e0c..e6e1e8915 100644
--- a/README.md
+++ b/README.md
@@ -158,7 +158,7 @@ agn connect my-aider <workspace-token>              # connect Aider into a works
 
 **Provider, model & key.** `AIDER_PROVIDER` decides which provider environment
 variable your `LLM_API_KEY` is injected into. It accepts `auto` (default),
-`openai`, `anthropic`, `openrouter`, `gemini`, `deepseek`, or
+`openai`, `anthropic`, `openrouter`, `requesty`, `gemini`, `deepseek`, or
 `openai-compatible`:
 
 | `AIDER_PROVIDER` | Key injected as | Notes |
@@ -166,6 +166,7 @@ variable your `LLM_API_KEY` is injected into. It accepts `auto` (default),
 | `anthropic` | `ANTHROPIC_API_KEY` | e.g. `AIDER_MODEL=sonnet` / `opus` / `claude-3-5-sonnet-20241022` |
 | `openai` | `OPENAI_API_KEY` | e.g. `AIDER_MODEL=gpt-4o` |
 | `openrouter` | `OPENROUTER_API_KEY` | e.g. `AIDER_MODEL=openrouter/anthropic/claude-3.5-sonnet` |
+| `requesty` | `REQUESTY_API_KEY` | e.g. `AIDER_MODEL=requesty/openai/gpt-4o` |
 | `gemini` | `GEMINI_API_KEY` | e.g. `AIDER_MODEL=gemini/gemini-1.5-pro` |
 | `deepseek` | `DEEPSEEK_API_KEY` | e.g. `AIDER_MODEL=deepseek/deepseek-chat` |
 | `openai-compatible` | `OPENAI_API_KEY` + `OPENAI_API_BASE` | **requires `LLM_BASE_URL`**; the model is normalized to `openai/<model>` |
diff --git a/llms.txt b/llms.txt
index 38b1392ea..89b5ffc61 100644
--- a/llms.txt
+++ b/llms.txt
@@ -579,6 +579,7 @@ This allows you to configure LLM settings once at the environment level rather t
 | `together` | `TOGETHER_API_KEY` | meta-llama/Llama-2-70b-chat-hf |
 | `perplexity` | `PERPLEXITY_API_KEY` | llama-3.1-sonar-huge-128k-online |
 | `openrouter` | `OPENROUTER_API_KEY` | (any model via OpenRouter) |
+| `requesty` | `REQUESTY_API_KEY` | (any model via Requesty) |
 | `bedrock` | `AWS_ACCESS_KEY_ID` | anthropic.claude-3-5-sonnet |
 | `custom` | `CUSTOM_API_KEY` | (requires DEFAULT_LLM_BASE_URL) |
 
diff --git a/sdk/src/openagents/config/llm_configs.py b/sdk/src/openagents/config/llm_configs.py
index 27c26f747..440c91c6c 100644
--- a/sdk/src/openagents/config/llm_configs.py
+++ b/sdk/src/openagents/config/llm_configs.py
@@ -27,6 +27,7 @@ class LLMProviderType(str, Enum):
     PERPLEXITY = "perplexity"
     GROQ = "groq"
     OPENROUTER = "openrouter"
+    REQUESTY = "requesty"
     MINIMAX = "minimax"
     LITELLM = "litellm"
     CUSTOM = "custom"  # Custom OpenAI-compatible endpoint
@@ -186,6 +187,13 @@ class LLMProviderType(str, Enum):
         "models": [],  # User specifies model name (e.g., "anthropic/claude-3-opus")
         "API_KEY_ENV_VAR": "OPENROUTER_API_KEY",
     },
+    # Requesty (OpenAI-compatible LLM gateway)
+    "requesty": {
+        "provider": "generic",
+        "api_base": "https://router.requesty.ai/v1",
+        "models": [],  # User specifies model name (e.g., "openai/gpt-4o-mini")
+        "API_KEY_ENV_VAR": "REQUESTY_API_KEY",
+    },
     # MiniMax
     "minimax": {
         "provider": "minimax",
@@ -495,6 +503,7 @@ def create_model_provider(
         "perplexity",
         "groq",
         "openrouter",
+        "requesty",
     ]:
         # Use predefined API base if not provided
         if not api_base and provider in MODEL_CONFIGS: