Krenar Emini comments

Results 9 comments of


                                            Krenar Emini

Run Ollama on a TPU

This may be interesting. Is Ollama officially supported on the TPU?

Fix crash on Ubuntu 22.04 (Python 3.10)

I had the same issue. This small correction seems to resolve this.

Rocm support planned

Any plans for ROCM support?

Real ram use is higher than listed in readme

I recently tested Moondream 0.5B on CPU only where it uses about 2,28 GiB of memory.

README.md changes

To many differences since last commit. File will be reviewed.

[ERROR] cuda device not available

My machine is running Linux Mint 21.3 x86_64 with an AMD Radeon RX 7900 XTX. Which config would be helpful?

[ERROR] cuda device not available

Perhaps I do misunderstand the description, but: The project description states that this project is cross-platform and runs across different compute backend (AMD GPU included).

Error: Failed to fetch completions: Error processing prompt (see logs with DEBUG>=2): [TODO] Unsupported model: model/huggingface/hub/models--TriAiExperiments--SFR-Iterative-DPO-LLaMA-3-8B-R/snapshots/347a2e393f548790707283151105ea15aebdd4e7

Any solution for this problem?

Komf Kavita Network Error

Any update on this?