[GH-ISSUE #3485] Whether Qwen from ollama library is Qwen1.5? #2146

New Issue

GiteaMirror · 2026-04-12T12:22:38-05:00

GiteaMirror commented

2026-04-12 12:22:38 -05:00

Originally created by @OPDEV001 on GitHub (Apr 4, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3485

What model would you like?

When I type ollama run qwen:7b and the command completed successfully, but may I know whether the Qwen is Qwen or Qwen1.5, because I saw Qwen1.5 in Github.

If not, please add Qwen1.5 to ollama library.

Thanks,

Originally created by @OPDEV001 on GitHub (Apr 4, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3485 ### What model would you like? When I type ollama run qwen:7b and the command completed successfully, but may I know whether the Qwen is Qwen or Qwen1.5, because I saw Qwen1.5 in Github. If not, please add Qwen1.5 to ollama library. Thanks,

GiteaMirror closed this issue

2026-04-12 12:22:38 -05:00

GiteaMirror commented

2026-04-12 12:22:39 -05:00

@onlyjokers commented on GitHub (Apr 4, 2024):

https://ollama.com/library/qwen:14b-chat-q4_0
looks like they're using Qwen 1.5.

@onlyjokers commented on GitHub (Apr 4, 2024): https://ollama.com/library/qwen:14b-chat-q4_0 looks like they're using Qwen 1.5.

GiteaMirror commented

2026-04-12 12:22:39 -05:00

@OPDEV001 commented on GitHub (Apr 5, 2024):

Hoping there has any exsiting methods to check the Qwen version.

I had to use the creation time of filesystem, and found it was created by several days, :). Based on the information from home page, the homepage said it was released at 8 weeks before, then it must be Qwen1.5.

Only one problem I found, that the performance is different, Qwen:14b from default ollama is better than Qwen1.5:7b-Chat on local python. No more time to test other modes, all tests completed on same machine.

Thanks Ollama Team, 👍

@OPDEV001 commented on GitHub (Apr 5, 2024): Hoping there has any exsiting methods to check the Qwen version. I had to use the creation time of filesystem, and found it was created by several days, :). Based on the information from home page, the homepage said it was released at 8 weeks before, then it must be Qwen1.5. Only one problem I found, that the performance is different, Qwen:14b from default ollama is better than Qwen1.5:7b-Chat on local python. No more time to test other modes, all tests completed on same machine. Thanks Ollama Team, 👍

GiteaMirror commented

2026-04-12 12:22:40 -05:00

@jmorganca commented on GitHub (Apr 17, 2024):

Hi there, sorry this isn't easier to inspect for downloaded models. The full list of qwen models is here https://ollama.com/library/qwen/tags

You can compare their ID to see which ones are the same - both 1 and 1.5 are there - the latest and 7b etc "tags" will default to the latest models (1.5)

@jmorganca commented on GitHub (Apr 17, 2024): Hi there, sorry this isn't easier to inspect for downloaded models. The full list of `qwen` models is here https://ollama.com/library/qwen/tags You can compare their ID to see which ones are the same - both 1 and 1.5 are there - the `latest` and `7b` etc "tags" will default to the latest models (1.5)

GiteaMirror commented

2026-04-12 12:22:40 -05:00

@OPDEV001 commented on GitHub (Apr 23, 2024):

Many thanks, I saw there has v1.5 including in tags.

I will update all my models in ollama, :)

@OPDEV001 commented on GitHub (Apr 23, 2024): Many thanks, I saw there has v1.5 including in tags. I will update all my models in ollama, :)