github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #4167] abnormal reply of Llama-3-ChatQA-1.5-8B-GGUF #2590

New Issue
Closed
opened 2026-04-12 12:56:27 -05:00 by GiteaMirror · 7 comments
GiteaMirror commented 2026-04-12 12:56:27 -05:00
Owner
Copy Link

Originally created by @taozhiyuai on GitHub (May 5, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4167

What is the issue?

import Llama-3-ChatQA-1.5-8B-GGUF to ollama, but reply is abnormal. I have tried many gguf version of this model from different username on HF.

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.1..32

Originally created by @taozhiyuai on GitHub (May 5, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4167 ### What is the issue? import Llama-3-ChatQA-1.5-8B-GGUF to ollama, but reply is abnormal. I have tried many gguf version of this model from different username on HF. ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.1..32
GiteaMirror added the bug label 2026-04-12 12:56:27 -05:00
GiteaMirror commented 2026-04-12 12:56:29 -05:00
Author
Owner
Copy Link

@DuckyBlender commented on GitHub (May 5, 2024):

Llama 3 ChatQA is a very wierd model which is optimised for RAG. It's still not 100% supported in ollama

<!-- gh-comment-id:2094902193 --> @DuckyBlender commented on GitHub (May 5, 2024): Llama 3 ChatQA is a very wierd model which is optimised for RAG. It's still not 100% supported in ollama
GiteaMirror commented 2026-04-12 12:56:30 -05:00
Author
Owner
Copy Link

@taozhiyuai commented on GitHub (May 5, 2024):

Llama 3 ChatQA is a very wierd model which is optimised for RAG. It's still not 100% supported in ollama

70B works. only 8B not. even I try 8B GGUF in LM Studio, it doesn't work.

https://ollama.com/taozhiyuai/llama3-chatqa-1.5-70b.q4_k_m

https://ollama.com/taozhiyuai/llama3-chatqa-1.5-70b.q8_0

<!-- gh-comment-id:2094979901 --> @taozhiyuai commented on GitHub (May 5, 2024): > Llama 3 ChatQA is a very wierd model which is optimised for RAG. It's still not 100% supported in ollama 70B works. only 8B not. even I try 8B GGUF in LM Studio, it doesn't work. https://ollama.com/taozhiyuai/llama3-chatqa-1.5-70b.q4_k_m https://ollama.com/taozhiyuai/llama3-chatqa-1.5-70b.q8_0
GiteaMirror commented 2026-04-12 12:56:31 -05:00
Author
Owner
Copy Link

@lazyracket commented on GitHub (May 9, 2024):

@taozhiyuai Take a look at this: https://github.com/ollama/ollama/issues/4101#issuecomment-2101187527
The .GGUF file is fine because it works well in LM Studio, so the Modelfile for Ollama is probably written incorrectly...

<!-- gh-comment-id:2101973874 --> @lazyracket commented on GitHub (May 9, 2024): @taozhiyuai Take a look at this: https://github.com/ollama/ollama/issues/4101#issuecomment-2101187527 The .GGUF file is fine because it works well in LM Studio, so the `Modelfile` for Ollama is probably written incorrectly...
GiteaMirror commented 2026-04-12 12:56:32 -05:00
Author
Owner
Copy Link

@thinkverse commented on GitHub (May 11, 2024):

Maybe try the version The Ollama team released a few hours ago - https://ollama.com/library/llama3-chatqa, as part of the 0.1.35 release.

<!-- gh-comment-id:2105505351 --> @thinkverse commented on GitHub (May 11, 2024): Maybe try the version The Ollama team released a few hours ago - https://ollama.com/library/llama3-chatqa, as part of the [0.1.35 release](https://github.com/ollama/ollama/releases/tag/v0.1.35).
GiteaMirror commented 2026-04-12 12:56:33 -05:00
Author
Owner
Copy Link

@lazyracket commented on GitHub (May 11, 2024):

Maybe try the version The Ollama team released a few hours ago - https://ollama.com/library/llama3-chatqa, as part of the 0.1.35 release.

Just tested it and it's working.

<!-- gh-comment-id:2105573531 --> @lazyracket commented on GitHub (May 11, 2024): > Maybe try the version The Ollama team released a few hours ago - https://ollama.com/library/llama3-chatqa, as part of the [0.1.35 release](https://github.com/ollama/ollama/releases/tag/v0.1.35). Just tested it and it's working.
GiteaMirror commented 2026-04-12 12:56:34 -05:00
Author
Owner
Copy Link

@Erik262 commented on GitHub (May 11, 2024):

Do you know which Embedding Model works best with llama3-chatqa? or in general? (5GB max model size)

<!-- gh-comment-id:2105594838 --> @Erik262 commented on GitHub (May 11, 2024): Do you know which Embedding Model works best with llama3-chatqa? or in general? (5GB max model size)
GiteaMirror commented 2026-04-12 12:56:35 -05:00
Author
Owner
Copy Link

@lazyracket commented on GitHub (May 11, 2024):

Do you know which Embedding Model works best with llama3-chatqa? or in general? (5GB max model size)

Also confused about this, testing it for dialog now works well enough, but how to apply it to RAG is more appropriate...

<!-- gh-comment-id:2105608307 --> @lazyracket commented on GitHub (May 11, 2024): > Do you know which Embedding Model works best with llama3-chatqa? or in general? (5GB max model size) Also confused about this, testing it for dialog now works well enough, but how to apply it to RAG is more appropriate...
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#2590
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?