Repository navigation
Ollama tag openbmb/minicpm5 defaults to top_p 0.8, matching neither documented sampling profile #360
Description
What
The openbmb/minicpm5 Ollama tag ships these defaults:
$ curl -s localhost:11434/api/show -d '{"model":"openbmb/minicpm5"}' | jq -r '.parameters'
top_p 0.8
stop "<|im_end|>"
stop "<|endoftext|>"
temperature 0.7
The MiniCPM5-1B model card documents top_p 0.95 for both sampling profiles (Think: temperature 0.9, top_p 0.95; No-Think: temperature 0.7, top_p 0.95).
References: ollama.com/openbmb/minicpm5
Why
temperature 0.7 is consistent with choosing the No-Think profile as the single static default — a reasonable choice, since a Modelfile carries one profile while think is toggleable per request. top_p 0.8 matches neither documented profile, so Ollama users sample outside both recommended configurations, and it is unclear which mode the defaults represent.
How
- If unintentional: update the tag's Modelfile to
top_p 0.95, aligning with the documented No-Think profile - If intentional: document the Ollama-specific default and its rationale on the tag page or deploy guide
- Optional: note in the tag description which profile (Think vs No-Think) the defaults target
Environment: Ollama server 0.30.5 / client 0.30.7, macOS (Apple Silicon), openbmb/minicpm5:latest (q4_K_M, 688 MB).
Activity
zhangtao2-1 commentedon Jun 16, 2026
@laurigates
thanks for the precise report — you're right, and this is an oversight on our side.
The intent was to ship the No-Think profile as the static default (hence temperature 0.7), since a Modelfile can only carry one profile while think is toggled per request. The top_p 0.8 doesn't match either documented profile — it's a leftover default and should be 0.95.
We'll update the openbmb/minicpm5 Ollama Modelfile to top_p 0.95 to align with the documented No-Think profile, and add a note on the tag page clarifying that the defaults target No-Think (with think toggleable per request). In the meantime, you can override per request, e.g. "options": {"top_p": 0.95}.
Will update here once the tag is re-pushed. Thanks again for flagging it.
laurigates commentedon Aug 19, 2026
This was closed as completed on 2026-07-09, but the Ollama tags still ship
top_p 0.8— the re-push described above doesn't appear to have happened.Checked today (2026-08-19) by reading the
application/vnd.ollama.image.paramsblob from each tag's manifest onregistry.ollama.ai:latest{"stop":["<|im_end|>","<|endoftext|>"],"temperature":0.7,"top_p":0.8}q4_K_Mq8_0fp16Same result from a client after a fresh pull:
So the state is unchanged from the original report:
temperature 0.7matches the documented No-Think profile,top_p 0.8matches neither profile (both are documented at0.95).Could this be reopened until the tags carry
top_p 0.95? Happy to re-verify once they're re-pushed.Not urgent for us — we override per request, as suggested. Flagging it because the closed state implies the defaults are now correct, and anyone taking the tag defaults at face value would still be sampling outside both documented profiles.