Akarshan Biswas

AMDGPU doesn't support radeon_powersave

7

Please use `echo high|low|auto in /sys/class/drm/card0/device/power_dpm_force_performance_level` tuned logs: ``` 2018-11-01 19:26:03,813 INFO tuned.daemon.application: dynamic tuning is enabled (can be overridden in plugins) ✔ 1.85G RAM 2018-11-01 19:26:03,820 INFO tuned.daemon.daemon: using...

feat: Custom OpenAI compatible server support with preserved inference parameters

4

**Problem** Sometimes, people like me run an openai compatible server instead of running the model through Jan's nitro for better tailored to hardware compatibility and availability. The openai inference partially...

Preference window do not seem to save the preferences

1

I am trying the local model option and providing it the local openAI server endpoint and it do not save the preferences (it doesn't even need a model name or...

Feat: Add support for Llama 3.1 function calling

3

Llama 3.1 models now officially supports function calling. So I think it would be nice to have function calling support for Llama 3.1 alongside existing functionary models. Looks like we...

I am trying to write a simple mamba model by studying the first mamba paper just for testing on a small dataset in JAX. Some tests I came across online...

SYCL: Fix GGML_SYCL_DEBUG macro

2

Originally, this work was part of #11515. Currently this macro doesn't work on files other than ggml-sycl.cpp because of a variable `g_ggml_sycl_debug` which gets initialized as a local variable only...