model
gpt-oss 20B llm
A model record: onboarded once for the project and used by any eval's arms.
- weights
- open
- licence
- Apache-2.0
- aliases
gpt-oss:20b- settings
reasoning_effortlow reasoning cannot be turned off: with none it still reasoned on every machine (onboarding, 2026-10-31); low is the least it takes- onboarding
- onboarded, no arm smoked 2026-10-31
- best result per goal
- #26 of 55 Triage decisions on Orrery's error signals
01deployments and their onboarding
x3 · x3 · ollama 0.40.13 · gpt-oss:20b · onboarding pass 2026-10-31 · 40.9 tokens/s · 1 accepted
- where
- x3
- server or device
- ollama 0.40.13
- build
gpt-oss:20bf38aa0c53da5f8c49d08c43a99df24ff53167fe68e24664a7777288e7656fdfeMXFP4- matches the record
- yes
- checked
- 2026-10-31
- onboarded on
- x3
| probe | measurement | flag |
|---|---|---|
| endpoint | provider openai · base url http://127.0.0.1:11434/v1 · model gpt-oss:20b | — |
| digest found | f38aa0c53da5f8c49d08c43a99df24ff53167fe68e24664a7777288e7656fdfe | — |
| default | s 6.97 · tokens out 107 · reasoned: yes · finish stop | — |
| chosen | s 1.15 · tokens out 47 · reasoned: yes · finish stop · settings (reasoning effort low) | — |
| tokens per s | 40.9 | — |
| deterministic | yes | — |
| json valid | yes | — |
Accepted: reasoning: it still reasons with the record's settings (it reasons whatever it is asked: with none and with low on every machine (onboarding, 2026-10-31); low is the least)
dgx · dgx · ollama 0.40.13 · gpt-oss:20b · onboarding pass 2026-10-31 · 52.3 tokens/s · 1 accepted
- where
- dgx
- server or device
- ollama 0.40.13
- build
gpt-oss:20bf38aa0c53da5f8c49d08c43a99df24ff53167fe68e24664a7777288e7656fdfe- matches the record
- yes
- checked
- 2026-10-31
- onboarded on
- dgx
| probe | measurement | flag |
|---|---|---|
| endpoint | provider openai · base url http://100.69.49.126:11434/v1 · model gpt-oss:20b | — |
| digest found | f38aa0c53da5f8c49d08c43a99df24ff53167fe68e24664a7777288e7656fdfe | — |
| default | s 8.36 · tokens out 127 · reasoned: yes · finish stop | — |
| chosen | s 0.9 · tokens out 47 · reasoned: yes · finish stop · settings (reasoning effort low) | — |
| tokens per s | 52.3 | — |
| deterministic | yes | — |
| json valid | yes | — |
Accepted: reasoning: it still reasons with the record's settings (it reasons whatever it is asked: with none and with low on every machine (onboarding, 2026-10-31); low is the least)
mac · mac · ollama 0.40.13 · gpt-oss:20b · onboarding pass 2026-10-31 · 40.7 tokens/s · 1 accepted
- where
- mac
- server or device
- ollama 0.40.13
- build
gpt-oss:20bf38aa0c53da5f8c49d08c43a99df24ff53167fe68e24664a7777288e7656fdfeMXFP4- matches the record
- yes
- checked
- 2026-10-31
- onboarded on
- mac
| probe | measurement | flag |
|---|---|---|
| endpoint | provider openai · base url http://127.0.0.1:11435/v1 · model gpt-oss:20b | — |
| digest found | f38aa0c53da5f8c49d08c43a99df24ff53167fe68e24664a7777288e7656fdfe | — |
| default | s 7.17 · tokens out 107 · reasoned: yes · finish stop | — |
| chosen | s 1.16 · tokens out 47 · reasoned: yes · finish stop · settings (reasoning effort low) | — |
| tokens per s | 40.7 | — |
| deterministic | yes | — |
| json valid | yes | — |
Accepted: reasoning: it still reasons with the record's settings (it reasons whatever it is asked: with none and with low on every machine (onboarding, 2026-10-31); low is the least)
What it can do, across its deployments: reasons: yes · json mode: yes · reasoning off 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 |
| 2026-10-31 | reasoning_effort: none → low | reasoning cannot be turned off: with none it still reasoned on every machine (onboarding, 2026-10-31); low is the least it takes |
03arms