Skip to content

Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.

C++

docs: STRATA_VCVARS, /metrics/prometheus behind the API key, and the Monitor's prompt speed (#3244)

Norme6f598c0a2a0
diff --git a/SECURITY.md b/SECURITY.md
index 8bd61c841..8afbf44cf 100644
--- a/SECURITY.md
+++ b/SECURITY.md
@@ -23,8 +23,9 @@ this PC only. The details and every setting are in [docs/DETAILS.md](docs/DETAIL
- **Where it listens.** `127.0.0.1` by default. `--host 0.0.0.0` (or `"host"` in `strata-<model>.json`) opens it to
your network, and the server then warns when no API key is set.
- **API key.** `"api_key"` in the run config (or `STRATA_API_KEY`) is required on `/v1/*` and on every endpoint
- that shows the model's state, requests or answers (`/status`, `/metrics`, `/settings`, `/mcp`, `/props`, `/slots`,
- `/api/requests`, `/config`) and on every `POST`. It is compared in constant time. Set one before you open the
+ that shows the model's state, requests or answers (`/status`, `/metrics`, `/metrics/prometheus`, `/settings`,
+ `/mcp`, `/props`, `/slots`, `/api/requests`, `/config`) and on every `POST`. It is compared in constant time.
+ Set one before you open the
server to your network or put a tunnel in front of it.
- **Host check (DNS rebinding, 0.1.38).** Without an API key the server answers only requests whose `Host` is a name
it knows (`localhost`, an IP address, the address it listens on and, when it listens beyond this PC, this PC's
diff --git a/docs/DETAILS.md b/docs/DETAILS.md
index 43538ddca..9ccf7685c 100644
--- a/docs/DETAILS.md
+++ b/docs/DETAILS.md
@@ -292,7 +292,9 @@ files, read time; `routing prefetch`: how many file reads had been warmed). The
`GET /metrics` lists `ram_blobs`, `file_blobs`, `file_mb`, `drafts_offered` and `drafts_accepted` per recent request, with
`totals` since the start. `hit_rate` is the VRAM share of the experts looked up; experts the GPU reads over PCIe
(`--pcie-frac`) are not in it, so a higher `--pcie-frac` raises it even when decoding gets slower. `pcie_share` is their
-share of all routed experts. The Monitor tab shows both.
+share of all routed experts. The Monitor tab shows both. The request list in the Monitor tab also shows the prompt's own speed: what the cache already held is not read
+again, so it is the fresh tokens (`prompt_tokens` minus `reused`) over the prompt time (`prompt_ms`), and a request that read no
+new token shows no number.
Profiling: start the server with `STRATA_DECODE_TIMING=1` (and `STRATA_VERIFY_PROFILE=1` for the GPU side). After each request
the engine log has a `strata decode timing:` line and a `strata decode GPU stages (ms/window):` line. The GPU profile slows
diff --git a/docs/SOURCE_INSTALL.md b/docs/SOURCE_INSTALL.md
index 1c0d7ec07..1db381508 100644
--- a/docs/SOURCE_INSTALL.md
+++ b/docs/SOURCE_INSTALL.md
@@ -34,6 +34,13 @@ In PowerShell, prefix the wrapper with `.\`, for example `.\START-HERE.bat`.
`SETUP.bat` opens setup again to add a model or change settings. The wrapper can reuse an existing install;
first-start and later load times depend on the pack, disk, RAM and backend. A model can take several minutes to load.
+When no ready-made engine fits the PC, setup compiles one. That needs the Visual Studio 2022 Build Tools with the C++ workload;
+Visual Studio 2019 is too old (the sources are C++20). With a CUDA toolkit, CUDA 13.0 to 13.2 accept Visual Studio 2022 only, and
+Visual Studio 2026 works from CUDA 13.3. The CPU image encoder alone (no CUDA toolkit on the PC) builds with MSVC and takes any
+Visual Studio that has the C++ tools. If setup cannot find an install it could use, name the file before running it, for example
+`set STRATA_VCVARS=C:\Program Files\Microsoft Visual Studio\2022\BuildTools\VC\Auxiliary\Build\vcvars64.bat`. Setup uses that file
+only if it exists; a wrong path is answered as no build tools at all.
+
## From source: Linux
Clone or extract the source, enter its folder, then run: