Ilker ManapandClaude Opus 5.5 86e9a1cb83 Give V100s the last vLLM built for them, not a build without their kernels ...

Read from vLLM's release pipeline: every x86 build from 0.20.0 has
kernels for compute 7.5 and up, the CUDA 12.9 one included, and 0.19.1
is the last built for 7.0. PyPI's build is CUDA 13 from 0.20, not 0.26.
A V100 asked for a newer vLLM is refused, naming 0.19.1 for
engine_overrides; the CUDA 12.9 fallback is for older drivers on
Turing and newer. The cu129 wheel is found among the release's assets,
since its manylinux tag differs between versions.

Refs #2.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

2026-10-01 14:04:45 +02:00

..

choosing-models.md

Manage people and the tenancy tree, from the web and the CLI

2026-09-28 16:26:39 +02:00

data.md

Answer the two questions asked before a gateway is allowed near anything

2026-09-27 08:16:08 +02:00

fabrics.md

Measure two nodes' fabric rail by rail: llmgw ctl fabric-test

2026-09-29 21:29:23 +02:00

facility-aware-placement.md

Mark facility steps 1-5 deferred; node sensors collected

2026-09-30 20:25:50 +02:00

install.md

Give V100s the last vLLM built for them, not a build without their kernels

2026-10-01 14:04:45 +02:00

node-onboarding.md

Add a node with one command: llmgw node add, with a terminal screen

2026-09-29 19:33:47 +02:00

onboarding.md

Document what changed since v0.9.2, and fabrics for the sites that have them

2026-09-29 10:14:58 +02:00

quick-install.md

Put the mirror on its own machine in two commands

2026-09-29 22:17:24 +02:00

quickstart.md

Add a quick install guide: six steps from empty servers to a model

2026-09-29 18:13:05 +02:00

reference.md

Decide engines per node, and let a node join without one it cannot run

2026-10-01 12:07:25 +02:00