model
Qwen 3.8 27B llm
A model record: onboarded once for the project and used by any eval's arms.
- weights
- open
- licence
- Apache-2.0
- aliases
qwen3.8:27b- 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
- #50 of 55 Triage decisions on Orrery's error signals
01deployments and their onboarding
x3 · x3 · ollama 0.40.13 · qwen3.8:27b · onboarding pass 2026-10-31 · 22.8 tokens/s
- where
- x3
- server or device
- ollama 0.40.13
- build
qwen3.8:27baaee06c39dcf2437cde036998d960e1fc1494b8191be7cc9657d01e509097813Q4_K_M- 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 qwen3.8:27b | — |
| digest found | aaee06c39dcf2437cde036998d960e1fc1494b8191be7cc9657d01e509097813 | — |
| default | s 8.66 · tokens out 79 · reasoned: yes · finish stop | — |
| chosen | s 1.01 · tokens out 23 · reasoned: no · finish stop · settings (reasoning effort none) | — |
| tokens per s | 22.8 | — |
| deterministic | yes | — |
| json valid | yes | — |
dgx · dgx · ollama 0.40.13 · qwen3.8:27b · onboarding pass 2026-10-31 · 23.7 tokens/s
- where
- dgx
- server or device
- ollama 0.40.13
- build
qwen3.8:27baaee06c39dcf2437cde036998d960e1fc1494b8191be7cc9657d01e509097813- 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 qwen3.8:27b | — |
| digest found | aaee06c39dcf2437cde036998d960e1fc1494b8191be7cc9657d01e509097813 | — |
| default | s 8.18 · tokens out 67 · reasoned: yes · finish stop | — |
| chosen | s 0.97 · tokens out 23 · reasoned: no · finish stop · settings (reasoning effort none) | — |
| tokens per s | 23.7 | — |
| deterministic | yes | — |
| json valid | yes | — |
mac · mac · ollama 0.40.13 · qwen3.8:27b · onboarding pass 2026-10-31 · 20.2 tokens/s
- where
- mac
- server or device
- ollama 0.40.13
- build
qwen3.8:27baaee06c39dcf2437cde036998d960e1fc1494b8191be7cc9657d01e509097813Q4_K_M- 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 qwen3.8:27b | — |
| digest found | aaee06c39dcf2437cde036998d960e1fc1494b8191be7cc9657d01e509097813 | — |
| default | s 16.75 · tokens out 73 · reasoned: yes · finish stop | — |
| chosen | s 1.14 · tokens out 23 · reasoned: no · finish stop · settings (reasoning effort none) | — |
| tokens per s | 20.2 | — |
| deterministic | yes | — |
| json valid | yes | — |
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