diff --git a/docs/my-website/docs/providers/hpc_ai.md b/docs/my-website/docs/providers/hpc_ai.md
index 4be00e9a7f4..79389d4d396 100644
--- a/docs/my-website/docs/providers/hpc_ai.md
+++ b/docs/my-website/docs/providers/hpc_ai.md
@@ -1,6 +1,6 @@
 # HPC-AI
 
-[HPC-AI](https://api.hpc-ai.com) provides an OpenAI-compatible inference API at `https://api.hpc-ai.com/inference/v1`.
+[HPC-AI](https://www.hpc-ai.com/) provides an OpenAI-compatible inference API at `https://api.hpc-ai.com/inference/v1`.
 
 :::tip
 
@@ -22,8 +22,6 @@ Optional: override the base URL (defaults to `https://api.hpc-ai.com/inference/v
 os.environ["HPC_AI_API_BASE"] = "https://api.hpc-ai.com/inference/v1"
 ```
 
-If you use another env name such as `HPC_AI_BASE_URL`, map it to `api_base` in your LiteLLM call or proxy `litellm_params`; LiteLLM reads `HPC_AI_API_BASE` by default.
-
 ## Sample Usage: Chat completion
 
 ```python
@@ -76,11 +74,18 @@ litellm --config /path/to/config.yaml
 
 3. Send requests to the proxy using your alias (`hpc-ai-minimax` in the example above).
 
-## Supported models (examples)
+## Supported models
 
 | LiteLLM model id | Notes |
 | ---------------- | ----- |
 | `hpc_ai/minimax/minimax-m2.5` | MiniMax M2.5 |
 | `hpc_ai/moonshotai/kimi-k2.5` | Kimi K2.5 |
 
-Pricing in `model_prices_and_context_window.json` may use placeholder token costs; set real rates when your billing API is available.
+## Pricing
+
+Costs in `model_prices_and_context_window.json` use USD per token. Approximate list rates (per 1M tokens):
+
+| Model | Uncached input | Cached input | Output |
+| ----- | -------------- | ------------ | ------ |
+| M2.5 (`minimax/minimax-m2.5`) | $0.30 | $0.03 | $1.20 |
+| K2.5 (`moonshotai/kimi-k2.5`) | $0.45 | $0.07 | $2.25 |
diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json
index 1758ea01481..fdac1253cb3 100644
--- a/model_prices_and_context_window.json
+++ b/model_prices_and_context_window.json
@@ -20189,26 +20189,30 @@
         ]
     },
     "hpc_ai/minimax/minimax-m2.5": {
-        "max_tokens": 262144,
-        "max_input_tokens": 262144,
-        "max_output_tokens": 262144,
-        "input_cost_per_token": 0.0,
-        "output_cost_per_token": 0.0,
+        "cache_read_input_token_cost": 3e-08,
+        "input_cost_per_token": 3e-07,
         "litellm_provider": "hpc_ai",
+        "max_input_tokens": 204800,
+        "max_output_tokens": 204800,
+        "max_tokens": 204800,
         "mode": "chat",
+        "output_cost_per_token": 1.2e-06,
+        "source": "https://api.hpc-ai.com/inference/v1",
         "supports_function_calling": true,
-        "source": "https://api.hpc-ai.com/inference/v1"
+        "supports_prompt_caching": true
     },
     "hpc_ai/moonshotai/kimi-k2.5": {
-        "max_tokens": 262144,
+        "cache_read_input_token_cost": 7e-08,
+        "input_cost_per_token": 4.5e-07,
+        "litellm_provider": "hpc_ai",
         "max_input_tokens": 262144,
         "max_output_tokens": 262144,
-        "input_cost_per_token": 0.0,
-        "output_cost_per_token": 0.0,
-        "litellm_provider": "hpc_ai",
+        "max_tokens": 262144,
         "mode": "chat",
+        "output_cost_per_token": 2.25e-06,
+        "source": "https://api.hpc-ai.com/inference/v1",
         "supports_function_calling": true,
-        "source": "https://api.hpc-ai.com/inference/v1"
+        "supports_prompt_caching": true
     },
     "hyperbolic/NousResearch/Hermes-3-Llama-3.1-70B": {
         "input_cost_per_token": 1.2e-07,