Skip to content

docs: replace the expired Discord invite on main (badge + site), no tag - #1472

Closed
JustVugg wants to merge 3 commits into
mainfrom
docs/discord-invite
Closed

JustVugg wants to merge 3 commits into
mainfrom
docs/discord-invite

Conversation

@JustVugg

Copy link
Copy Markdown
Owner

Same single commit as #1471, aimed at main so the badge and the community line on the default branch stop pointing at a dead link. Docs only, no tag: no release build runs and no archives change.

Not a dev -> main merge on purpose: dev is already ahead of v1.11.0 with #1468 (engine code), which must not reach main untagged. This carries exactly the ten replaced occurrences.

Verified live before opening: GET /api/v10/invites/RXV83nSZdk -> guild "Colibri - inference AI", expires_at: null. The old FkyrEeJR answers Invite is expired (50270).

Merge with --admin (main is protected).

JustVugg and others added 3 commits September 13, 2026 14:59
… in MoE"

#1464: an RTX 3080 under WSL2 failed every prompt with

    request failed: hybrid batched block failed in MoE

and a day of flag-flipping could not narrow it, because that message is the
generic tail of the layer block. moe_token_pipeline and v4_moe_batch_union
can fail in some thirty places -- an expert that would not read, an upload
that would not land, the GPU expert group refusing, a routing table missing,
a plain malloc -- and every one of them returned a bare -1. The block could
not tell an out-of-VRAM card from a bad read over /mnt/c, so neither could
the reporter.

Every failure site now records a reason (layer, expert where there is one,
and the step) through moe_fail(), thread-local and cleared on entry so a
stale one cannot outlive the call that set it, and both block tails append
it:

    hybrid batched block failed in MoE: layer 12: the GPU expert group of 6
    experts refused (a CUDA allocation or launch failed; check nvidia-smi
    for free VRAM)

    block computation failed in MoE: layer 3: reading expert 137 from the
    expert store failed

No behaviour changes on any success path; the only difference is the text
in the error buffer. Everything lives in the COLI_V4_UNIT_BLOCK_HYBRID unit,
which needed <stdarg.h> of its own since set_error's unit is a different
object.

tests/test_v4_moe_reason_source.py pins the shape: the only `result = -1`
left in those two functions are the "not finished yet" job markers, both
functions clear the reason on entry, and both block tails read it. It fails
on the tree before this commit. The tiny V4 oracle stays token-exact
(target long, CLI, serve, prefix reuse).
fix(deepseek-v4): the MoE step says why it failed, instead of "failed in MoE"
The invite we publish is dead again. The Discord API on the code in the tree:

    GET /api/v10/invites/FkyrEeJR  ->  {"message": "Invite is expired.", "code": 50270}

It appears in ten places: the header badge and the community line of all
four READMEs, and twice in site/index.html. Anyone clicking Discord from the
repo or the site lands on an error page.

Replaced with an invite created with no expiry, verified live before this
commit:

    GET /api/v10/invites/RXV83nSZdk  ->  guild "Colibri - inference AI",
                                         expires_at: null, ~130 members

The previous replacement (#1363) died because Discord's default invite
lifetime is seven days; the first candidate for this one had
expires_at 2026-09-20 and was rejected for the same reason.
@JustVugg

Copy link
Copy Markdown
Owner Author

Closing: this branch was cut from dev after #1468 had landed there, so against main it carries #1468's engine changes (c/deepseek_v4.c, a new test) as well as the invite, which is exactly what a docs-only merge must not do. Replaced by #1473, cut from main, one commit, five files, ten lines.

@JustVugg JustVugg closed this Sep 13, 2026
@JustVugg
JustVugg deleted the docs/discord-invite branch September 24, 2026 22:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant