#35229 [Bug]: Snowflake Claude provider drops cache_creation_input_tokens and cache_read_input_tokens from response usage
### Check for existing issues
- [x] I have searched the existing issues and checked that my issue is not a duplicate.
### What happened?
When using Snowflake as the provider, the usage field in the response does not contain cache-related fields.
It was found when I use litellm proxy, but the problem lies in the SDK.
The codes in `litellm/llms/snowflake/chat/transformation.py` 401-406 prove that the problem exists.
``` usage_data = response_json.get("usage", {}) usage = Usage( prompt_tokens=usage_data.get("input_tokens", 0), completion_tokens=usage_data.get("output_tokens", 0), total_tokens=usage_data.get("input_tokens", 0) + usage_data.get("output_tokens", 0), ) ```
### Steps to Reproduce
1. Use the configs for snowflake
``` - model_name: sf-claude-opus-4-5 litellm_params: model: snowflake/claude-opus-4-5 api_base: https://xxxx.snowflakecomputing.com/api/v2 api_key: pat/xxxx ```
2. Use the request body for proxy
``` { "cache_control": { "type": "ephemeral" }, "model": "sf-claude-opus-5", "max_tokens": 256, "messages": [ { "role": "user", "content": "longtext" } ], } ```
3. The responses usa…