model
DeepSeek 4.1 Flash llm
A model record: onboarded once for the project and used by any eval's arms.
- weights
- open
- licence
- MIT
- aliases
eval-deepseek-41flash- settings
reasoning_effortnone it reasoned by default on the onboarding probe (2026-10-31); off unless a task needs it (RUNBOOK §16): left at the server's default, one model reasons at length and the next not at all- onboarding
- vetted 2026-10-31
- best result per goal
- #2 of 29 Short news summaries (CNN)
#3 of 19 Evidence retrieval for scientific claims (SciFact)
#10 of 33 A scorer that agrees with experts on summaries (SummEval)
#11 of 28 Named-entity extraction (Few-NERD)
#12 of 28 Encyclopedia topic classification (DBpedia)
#23 of 30 News topic classification (AG News)
#47 of 55 Triage decisions on Orrery's error signals
01deployments and their onboarding
hosted · homelab LiteLLM (the fleet's gateway) · eval-deepseek-41flash · onboarding pass 2026-10-31 · 36.2 tokens/s · 2 accepted
- where
- homelab LiteLLM (the fleet's gateway)
- build
eval-deepseek-41flash- matches the record
- yes
- checked
- 2026-10-31
- onboarded on
- homelab LiteLLM (the fleet's gateway)
| probe | measurement | flag |
|---|---|---|
| endpoint | provider litellm · base url http://homelab:4001/v1 · model eval-deepseek-41flash | — |
| digest found | None | — |
| default | s 1.56 · tokens out 43 · reasoned: yes · finish stop | — |
| chosen | s 0.66 · tokens out 24 · reasoned: no · finish stop · settings (reasoning effort none) | — |
| tokens per s | 36.2 | — |
| deterministic | yes | — |
| json valid | yes | — |
Accepted: where it ran: a hosted gateway serves the provider's current build; there is no digest, revision or hash to pin; determinism: a hosted provider batches requests, so temperature 0 does not repeat exactly; repeats measure it
What it can do, across its deployments: reasons: yes · json mode: yes · reasoning off reasoning_effort: none
02settings history
| when | what changed | why |
|---|---|---|
| 2026-10-31 | reasoning_effort: (server default) → none | it reasoned by default on the onboarding probe (2026-10-31); off unless a task needs it (RUNBOOK §16): left at the server's default, one model reasons at length and the next not at all |
03arms