eval-harness

model

Mistral Medium 3.5 llm

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

weights
open
aliases
mistral-medium-3.5:128b eval-mistral-medium-35
settings
defaults
onboarding
onboarded, no arm smoked 2026-10-31
best result per goal
#44 of 55 Triage decisions on Orrery's error signals

Every model

01deployments and their onboarding

x3 · x3 · ollama 0.40.13 · mistral-medium-3.5:128b · onboarding pass 2026-10-31 · 2.0 tokens/s
where
x3
server or device
ollama 0.40.13
build
mistral-medium-3.5:128b
0341632adb051badc332bd814779ab931ab78619750bebe3b7c7b1401e9d70df
Q4_K_M
matches the record
yes
checked
2026-10-31
onboarded on
x3
probemeasurementflag
endpointprovider openai · base url http://127.0.0.1:11434/v1 · model mistral-medium-3.5:128b—
digest found0341632adb051badc332bd814779ab931ab78619750bebe3b7c7b1401e9d70df—
defaults 70.11 · tokens out 25 · reasoned: no · finish stop—
chosens 12.32 · tokens out 25 · reasoned: no · finish stop · settings ()—
tokens per s2.0—
deterministicyes—
json validyes—
dgx · dgx · ollama 0.40.13 · mistral-medium-3.5:128b · onboarding pass 2026-10-31 · 2.9 tokens/s
where
dgx
server or device
ollama 0.40.13
build
mistral-medium-3.5:128b
0341632adb051badc332bd814779ab931ab78619750bebe3b7c7b1401e9d70df
matches the record
yes
checked
2026-10-31
onboarded on
dgx
probemeasurementflag
endpointprovider openai · base url http://100.69.49.126:11434/v1 · model mistral-medium-3.5:128b—
digest found0341632adb051badc332bd814779ab931ab78619750bebe3b7c7b1401e9d70df—
defaults 29.51 · tokens out 25 · reasoned: no · finish stop—
chosens 8.7 · tokens out 25 · reasoned: no · finish stop · settings ()—
tokens per s2.9—
deterministicyes—
json validyes—
hosted · homelab LiteLLM (the fleet's gateway) · eval-mistral-medium-35 · onboarding pass 2026-10-31 · 34.5 tokens/s
where
homelab LiteLLM (the fleet's gateway)
build
eval-mistral-medium-35
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-mistral-medium-35—
digest foundNone—
defaults 0.84 · tokens out 25 · reasoned: no · finish stop—
chosens 0.73 · tokens out 25 · reasoned: no · finish stop · settings ()—
tokens per s34.5—
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.