[GH-ISSUE #5184] ollama show should have the exact parameter count rounded to 3 digits #49771

Open
opened 2026-04-28 12:53:39 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @jmorganca on GitHub (Jun 20, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5184

What is the issue?

% ollama show llama3
  Model                                              
  	arch            	llama	                              
  	parameters      	8.0B 	                              
  	quantization    	Q4_0 	                              
  	context length  	8192 	                              
  	embedding length	4096 	         

should be the same as

Screenshot 2024-06-20 at 10 21 28 AM

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @jmorganca on GitHub (Jun 20, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5184 ### What is the issue? ``` % ollama show llama3 Model arch llama parameters 8.0B quantization Q4_0 context length 8192 embedding length 4096 ``` should be the same as <img width="793" alt="Screenshot 2024-06-20 at 10 21 28 AM" src="https://github.com/ollama/ollama/assets/251292/3e9ac431-e205-45b9-91ef-b63ddc8fe0f9"> ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bugapi labels 2026-04-28 12:53:39 -05:00
Author
Owner

@royjhan commented on GitHub (Jun 20, 2024):

The reason for this is in this function:
Screenshot 2024-06-20 at 8 34 17 AM

So api/show is also returning 8.0B. This is before any of the new api and cli changes. Do we want to still have the api return 8.0B or do we want to change precision so it returns 8.03B?

Not clear to me why M has greater precision than B but I may be unaware of the reason

<!-- gh-comment-id:2180998238 --> @royjhan commented on GitHub (Jun 20, 2024): The reason for this is in this function: <img width="734" alt="Screenshot 2024-06-20 at 8 34 17 AM" src="https://github.com/ollama/ollama/assets/65097070/1aee33c3-0b4f-4367-b904-2b76c4572766"> So api/show is also returning 8.0B. This is before any of the new api and cli changes. Do we want to still have the api return 8.0B or do we want to change precision so it returns 8.03B? Not clear to me why M has greater precision than B but I may be unaware of the reason
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#49771