### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

Three Claude Haiku 4.5 entries in `model_prices_and_context_window.json` have wrong `max_output_tokens` / `max_tokens` values. Claude Haiku 4.5's maximum output is **64,000 tokens** (200K context) on every provider:

| Catalog entry | Current value | Correct value | Evidence | |---|---|---|---| | `vertex_ai/claude-haiku-4-5` | `max_tokens` / `max_output_tokens`: **8192** | **64000** | [Google Vertex AI docs](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/haiku-4-5): "Maximum output tokens: 64,000" | | `vertex_ai/claude-haiku-4-5@20251001` | `max_tokens` / `max_output_tokens`: **8192** | **64000** | same | | `openrouter/anthropic/claude-haiku-4.5` | `max_tokens` / `max_output_tokens`: **200000** | **64000** | OpenRouter API (`GET /api/v1/models`): `top_provider.max_completion_tokens: 64000` |

The 8192 looks carried over from Claude 3.5 Haiku's old output cap; the 200000 looks like the context window pasted into the output fields. For reference, litellm's own `claude-haiku-4-5` (anthropic) and `anthropic.claude-…