eval-harness

model

Claude Sonnet 5.5 llm

A model record: onboarded once for the project and used by any eval's arms.

weights
proprietary
aliases
eval-claude-sonnet-55
settings
defaults
onboarding
onboarded, no arm smoked 2026-10-31
best result per goal
#28 of 55 Triage decisions on Orrery's error signals

Every model

01deployments and their onboarding

hosted · homelab LiteLLM (the fleet's gateway) · eval-claude-sonnet-55 · onboarding pass 2026-10-31 · 44.9 tokens/s
where
homelab LiteLLM (the fleet's gateway)
build
eval-claude-sonnet-55
matches the record
yes
checked
2026-10-31
onboarded on
homelab LiteLLM (the fleet's gateway)
probemeasurementflag
endpointprovider litellm · base url http://homelab:4001/v1 · model eval-claude-sonnet-55—
digest foundNone—
defaults 2.09 · tokens out 61 · reasoned: no · finish stop—
chosens 1.42 · tokens out 64 · reasoned: no · finish stop · settings ()—
tokens per s44.9—
deterministicyes—
json validyes—

What it can do, across its deployments: reasons: no · json mode: yes

02settings history

No change of settings or build is recorded.

03arms

armdeploymentgoalits settingsarm smokeresult
triage_sonnet_55_v1 · rescored 5f2c057c
orrery_triage_v1_labelled
hostedTriage decisions on Orrery's error signals—nonetriage_cost 0.708