synartesis 0.8.6 → 0.8.8
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +46 -0
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,52 @@
|
|
|
2
2
|
|
|
3
3
|
What changed, and why it mattered. Dates are release dates.
|
|
4
4
|
|
|
5
|
+
## 0.8.8 — 2026-09-22
|
|
6
|
+
|
|
7
|
+
### Fixed
|
|
8
|
+
|
|
9
|
+
- **The GitHub Models preset added in 0.8.7 could never have worked.** GitHub
|
|
10
|
+
retired its inference API on 30 July 2026, three weeks before that entry was
|
|
11
|
+
written. The host is still there and still answers -- `200 OK`, with
|
|
12
|
+
`content-type: text/plain` and a body of `OK`, whatever path you ask for --
|
|
13
|
+
which is exactly the shape of thing that looks alive until you try to use
|
|
14
|
+
it. The entry was built from blog posts and community threads, every one of
|
|
15
|
+
them written before the retirement, while the vendor's own documentation
|
|
16
|
+
said in plain words that the API was gone.
|
|
17
|
+
|
|
18
|
+
Removed, and replaced with **OpenRouter**: one key, many models, and
|
|
19
|
+
nineteen of the free ones currently advertise tool support, which is the
|
|
20
|
+
constraint that actually decides whether a model can drive this window at
|
|
21
|
+
all. Its catalogue is public, so which nineteen can be checked rather than
|
|
22
|
+
guessed.
|
|
23
|
+
|
|
24
|
+
Groq, added in the same release, was checked the same way and is fine.
|
|
25
|
+
|
|
26
|
+
The note now at the head of the preset list is the part worth keeping: send
|
|
27
|
+
a request to the host before adding one. A preset that does not work is
|
|
28
|
+
worse than no preset, because the person who picks it concludes this window
|
|
29
|
+
is broken rather than that the service is.
|
|
30
|
+
|
|
31
|
+
## 0.8.7 — 2026-09-22
|
|
32
|
+
|
|
33
|
+
### Added
|
|
34
|
+
|
|
35
|
+
- **Two presets for people with no GPU.** The window has run free models since
|
|
36
|
+
it had a model picker -- Ollama, LM Studio and vLLM all need no key and
|
|
37
|
+
nothing leaves the machine -- but every one of those is free only if you own
|
|
38
|
+
the hardware to run it. **Groq** and **GitHub Models** are hosted, both have
|
|
39
|
+
a free tier, and neither wants a card: a GitHub token you already have is
|
|
40
|
+
enough for the second. Both speak `/v1/chat/completions`, so they are
|
|
41
|
+
presets rather than code, which is what the note at the head of that list
|
|
42
|
+
has always said any endpoint can be.
|
|
43
|
+
|
|
44
|
+
The note beside each says what it actually costs -- rate limited, and meant
|
|
45
|
+
for trying things rather than for a day's work -- because the failure worth
|
|
46
|
+
preventing here is somebody picking a free option and reading its quota
|
|
47
|
+
refusal as this window being broken. The same reason the local entries say
|
|
48
|
+
to pick a model that can call tools: the window is an MCP client, and a
|
|
49
|
+
model that cannot emit a tool call cannot drive it at all.
|
|
50
|
+
|
|
5
51
|
## 0.8.6 — 2026-09-22
|
|
6
52
|
|
|
7
53
|
A test I added in 0.8.5 failed the release build. The published package was
|