llmgw/internal at 89435735dea60c5284df5fa35deff034a61e575f - llmgw - Gitea: Mynodes.xyz

deniz/llmgw

0 Fork 0

forked from ilker/llmgw

Files

T

History

Ilker ManapandClaude Opus 5.5 a583359ebf Agents started together measure their memory apart

Seven VMs on one host restarted at once read a third less each than any
gets alone. An agent now waits 10 s to 2 min, at random, before measuring,
and measures again every hour, keeping the best.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-28 09:56:02 +01:00

..

People sign in with their Active Directory account

2026-10-18 10:59:50 +02:00

A cached prompt costs the cached rate

2026-10-25 17:15:48 +01:00

Agents started together measure their memory apart

2026-10-28 09:56:02 +01:00

Guardrails turn off and on; a judge's refusal says which hazard; ctl guard judge ask

2026-10-06 12:41:08 +02:00

Anthropic's Messages API: /v1/messages translated to chat completions and back, streamed or not, with tools and images; keys in x-api-key

2026-10-04 22:59:55 +02:00

ctl bench reads a guarded engine's metrics with the engine's key, or says why it cannot

2026-10-05 00:20:25 +02:00

Offline sites and Kubernetes: llmgw bundle create, import and trust, offline.mirror, server init --bundle; a Helm chart for the control plane

2026-10-05 00:13:39 +02:00

Canary routing: a model's canary takes its share of requests, a session stays on one side, x-llmgw-variant pins one, and the two are compared

2026-10-04 19:56:16 +02:00

Answers to similar questions, not only the same one

2026-10-25 17:47:11 +01:00

Pictures, 3D models, a vision model and workflows as tools at /mcp

2026-10-14 13:11:09 +02:00

Believe X-Forwarded-* only from router.trusted_proxies

2026-10-04 10:29:02 +02:00

ComfyUI models of one environment share one replica

2026-10-14 13:55:34 +02:00

Answers to similar questions, not only the same one

2026-10-25 17:47:11 +01:00

The memory is measured after the agent reaches its control plane

2026-10-27 15:31:21 +01:00

Sign-in is set up from the console, and a directory tried before it is kept

2026-10-19 12:55:27 +02:00

Weights that run code are refused

2026-10-25 23:17:22 +01:00

A node's own readiness and its multi-node readiness are kept apart

2026-10-07 13:11:16 +02:00

The firewall setting reaches the mirror machine and the store's members

2026-10-09 08:53:17 +02:00

A model that does not fit its card is not added to fail; weights in use are not removed

2026-10-15 00:22:01 +02:00

A guardrail policy can ask someone else's service

2026-10-25 16:49:56 +01:00

The explorer lists what this cluster's hardware holds

2026-10-26 18:30:16 +01:00

Bound idle time, header size and request reading on every listener, and connections on the client port

2026-10-04 10:21:15 +02:00

llmgw-kb collections names each collection's owner as people know it

2026-10-08 15:42:40 +02:00

Precise KV-index routing: vLLM KV cache events through the agent, routed by leading blocks held

2026-10-05 01:21:11 +02:00

A forgotten password is set through a one-time link

2026-10-07 13:20:11 +02:00

An MCP client for the company's tool servers

2026-10-20 10:43:35 +02:00

The firewall setting reaches the mirror machine and the store's members

2026-10-09 08:53:17 +02:00

How fast a model would answer on the cards, as this cluster would lay it out

2026-10-28 00:19:36 +01:00

Several keys for one outside provider

2026-10-17 08:56:30 +02:00

Tracing beyond the router: reconcile passes, placement and a replica's start traced down to the node

2026-10-05 00:03:17 +02:00

A node's own readiness and its multi-node readiness are kept apart

2026-10-07 13:11:16 +02:00

A lost store machine is given up on in seconds, and a new leader waits for its judge

2026-10-15 20:33:13 +02:00

A machine too small for the models it would serve is said so first

2026-10-25 16:19:00 +01:00

Show usage over time, per organization, team, project and person

2026-09-28 08:11:31 +02:00

Offline sites and Kubernetes: llmgw bundle create, import and trust, offline.mirror, server init --bundle; a Helm chart for the control plane

2026-10-05 00:13:39 +02:00

Answers to similar questions, not only the same one

2026-10-25 17:47:11 +01:00

gofmt what the security work left unformatted

2026-10-04 12:22:51 +02:00

Match a mapped project however its path is spaced: the store writes "Acme / ml / train"

2026-10-04 16:45:47 +02:00

Sign-in is set up from the console, and a directory tried before it is kept

2026-10-19 12:55:27 +02:00

The cluster's settings live in the store, and every change takes effect

2026-10-05 15:25:23 +02:00

Onboard a node with a script, and find two examples that never worked

2026-09-16 12:28:36 +02:00

Keep the control plane's fetches off link-local addresses, and see through NAT64, 6to4 and proxies

2026-10-04 10:48:01 +02:00

Answers to similar questions, not only the same one

2026-10-25 17:47:11 +01:00

Put etcd and Patroni's API on TLS, and move a running cluster there in maintenance mode

2026-10-04 12:07:17 +02:00

Switch the project language to English

2026-09-16 01:15:21 +02:00

VS Code gets the gateway's models from an extension the gateway serves

2026-10-24 01:39:50 +02:00

How fast a model would answer on the cards, as this cluster would lay it out

2026-10-28 00:19:36 +01:00