> 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/faq.md).

# FAQ

### Can I run the Spec Lab against OpenAI / Anthropic / Groq?

No — see [Honesty Gate](/seekspeed-docs/spec-lab/honesty-gate.md). You cannot meaningfully measure speculative decoding on a closed-weight hosted API because you can't control or even observe the inference stack.

### Does SeekSpeed train any models?

No. It is a benchmarking, profiling, and recommendation tool. The Spec Lab adapter reads metrics from your already-deployed inference server.

### How is this different from running `ab` or `wrk`?

`ab` and `wrk` measure HTTP throughput. They have no notion of TTFT, tokens, cost, draft acceptance, or A/B significance. SeekSpeed treats the LLM-specific signals as first-class.

### Why TanStack Start instead of Next.js / Remix?

We needed typed RPC (`createServerFn`) where the same module owns both the client signature and the server handler, with no separate API project. TanStack Start was the cleanest fit and ships with file-based routing on top of Vite 7.

### What's the minimum sample size for a significant A/B?

The UI hides the significance badge under `n < 5` per group. Realistically, plan for 20-30 iterations per group to get tight enough variance for `p < 0.05` to be meaningful.

### Can I bring my own optimizer rules?

The rule list in `src/lib/recommendations.ts` is plain TypeScript. Each rule is a function that takes a run summary and returns zero or more `Recommendation` objects with an `apply()` mutator. Add one, rebuild, ship.

### Where does the DeepSpec / DSpark integration live in code?

Adapter probes are in `src/lib/spec-lab.functions.ts`. Bottleneck math is in `src/lib/spec-bottlenecks.ts`. The UI panel is `src/components/spec-lab/BottleneckPanel.tsx`. The DeepSpec serving recipe and DSpark configuration are documented in [Spec Lab → DeepSpec](/seekspeed-docs/spec-lab/deepspec.md) and [Spec Lab → DSpark](/seekspeed-docs/spec-lab/dspark.md).

### Where do I report bugs?

Open an issue on the project's GitHub repo. Include the report JSON export if you can — it has everything we need to reproduce.


---

# 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/faq.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.