[GH-ISSUE #9222] The following model is poorly supported #68068

Closed
opened 2026-05-04 12:25:46 -05:00 by GiteaMirror · 4 comments
Owner

Originally created by @17Reset on GitHub (Feb 19, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9222

The following two models, quantised into gguf format using llamacpp, are created into ollama, and ollama's reasoning is severely hallucinated and never stops
https://huggingface.co/open-thoughts/OpenThinker-32B
https://huggingface.co/huihui-ai/OpenThinker-32B-abliterated

Originally created by @17Reset on GitHub (Feb 19, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9222 The following two models, quantised into gguf format using llamacpp, are created into ollama, and ollama's reasoning is severely hallucinated and never stops `https://huggingface.co/open-thoughts/OpenThinker-32B` `https://huggingface.co/huihui-ai/OpenThinker-32B-abliterated`
GiteaMirror added the model label 2026-05-04 12:25:46 -05:00
Author
Owner

@rick-github commented on GitHub (Feb 19, 2025):

How did you import the model?

<!-- gh-comment-id:2668753565 --> @rick-github commented on GitHub (Feb 19, 2025): How did you import the model?
Author
Owner

@17Reset commented on GitHub (Feb 19, 2025):

  1. Generate SHA256
    HASH=$(sha256sum "$file" | awk '{print $1}')
    2.Push File to Ollama Server
    curl -T "OpenThinker-32B/OpenThinker_32B_Q8_Abliterated.gguf" -X POST "http://0.0.0.0:11434/api/blobs/sha256:${HASH}"
  2. Create Model from GGUF File
    curl http://0.0.0.0:11434/api/create -d "{ \"model\": \"OpenThinker_32B_Q8\", \"files\": { \"OpenThinker-32B/OpenThinker_32B_Q8_Abliterated.gguf\": \"sha256:${HASH}\" } }"
<!-- gh-comment-id:2668828084 --> @17Reset commented on GitHub (Feb 19, 2025): 1. Generate SHA256 `HASH=$(sha256sum "$file" | awk '{print $1}')` 2.Push File to Ollama Server `curl -T "OpenThinker-32B/OpenThinker_32B_Q8_Abliterated.gguf" -X POST "http://0.0.0.0:11434/api/blobs/sha256:${HASH}"` 3. Create Model from GGUF File `curl http://0.0.0.0:11434/api/create -d "{ \"model\": \"OpenThinker_32B_Q8\", \"files\": { \"OpenThinker-32B/OpenThinker_32B_Q8_Abliterated.gguf\": \"sha256:${HASH}\" } }"`
Author
Owner

@rick-github commented on GitHub (Feb 19, 2025):

You need a template and a system message.

https://ollama.com/huihui_ai/openthinker-abliterated:32b-q8_0

<!-- gh-comment-id:2668841018 --> @rick-github commented on GitHub (Feb 19, 2025): You need a template and a system message. https://ollama.com/huihui_ai/openthinker-abliterated:32b-q8_0
Author
Owner

@17Reset commented on GitHub (Feb 21, 2025):

Thanks, it runs great after setting it up!

<!-- gh-comment-id:2673189031 --> @17Reset commented on GitHub (Feb 21, 2025): Thanks, it runs great after setting it up!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#68068