> For the complete documentation index, see [llms.txt](https://seekspeed.gitbook.io/seekspeed-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://seekspeed.gitbook.io/seekspeed-docs/reference/glossary.md).

# Glossary

| Term                              | Definition                                                                                           |
| --------------------------------- | ---------------------------------------------------------------------------------------------------- |
| **TTFT**                          | Time to first token. ms from request send to first non-empty SSE delta.                              |
| **Inter-token latency**           | mean ms between successive SSE deltas during decode.                                                 |
| **Throughput**                    | tokens out per wall-clock second for a single request.                                               |
| **p95 / p99**                     | latency percentiles across all iterations of a run.                                                  |
| **Welch's *****t*****-test** | unequal-variance two-sample *t*-test used for A/B comparisons.                                       |
| **Speculative decoding**          | inference acceleration where a draft model proposes *k* tokens that the target verifies in one pass. |
| **Draft model**                   | small fast model that proposes tokens.                                                               |
| **Target model**                  | large accurate model that verifies / rejects drafted tokens.                                         |
| **Acceptance rate (α)**           | fraction of drafted tokens the target accepts.                                                       |
| **Tokens per step**               | average tokens emitted per target forward pass under speculative decoding.                           |
| **DeepSpec**                      | open reference toolkit for speculative decoding on DeepSeek models.                                  |
| **DSpark**                        | DeepSeek's production speculative-decoding stack with quantised draft + dynamic-*k* scheduling.      |
| **Honesty Gate**                  | UX + server check that blocks Spec Lab runs against closed-weight hosted APIs.                       |
| **Workspace**                     | top-level tenant; owner-scoped via RLS.                                                              |
| **Agent profile**                 | unit under test: prompt + connector + tools + memory + budget.                                       |
| **Variant**                       | child agent generated by an applied recommendation.                                                  |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://seekspeed.gitbook.io/seekspeed-docs/reference/glossary.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.