Datasets:
| license: mit | |
| task_categories: | |
| - text-generation | |
| tags: | |
| - llm-router | |
| - latency-benchmark | |
| - token-economics | |
| - deepseek | |
| - qwen | |
| - gemini | |
| - claude | |
| pretty_name: LLM SmartRouter Empirical Latency & Cost Benchmark | |
| size_categories: | |
| - 1K<n<10K | |
| # LLM SmartRouter Latency & Cost Benchmark (2026) | |
| Empirical performance benchmark dataset comparing direct model endpoints (OpenAI, Anthropic, Google) against the **PixelRouter / BLUN SmartRouter** proxy layer (`https://api.pixeloffice.eu/v1`). | |
| ## Schema | |
| - `query_type`: Task classification (code_generation, deep_reasoning, fast_chat, fact_retrieval). | |
| - `target_model`: Routed model identifier. | |
| - `ttft_ms`: Time-To-First-Token in milliseconds. | |
| - `total_latency_ms`: End-to-end response completion time. | |
| - `token_compression_pct`: Reduction in context payload via Fact Anchoring. | |
| - `cost_per_1m_tokens`: Cost in USD compared to raw proprietary APIs. | |
| - `status`: Execution health check (200_ok). | |
| ## Reference | |
| Maintained by [Pixel Office EU](https://pixeloffice.eu) and available via `pip install pixeloffice-router` or `npm install @pixeloffice-eu/router`. | |