llama runner process has terminated
I keep getting this error when i try running the model using ollama
Error: llama runner process has terminated: GGML_ASSERT(hparams.n_swa > 0) failed
any idea what the issue is?
Ya lo tienes dentro de la web de ollama, no hace falta descargar desde aquí el gguf.
You're running an older llama.cpp without Phi 4 support. You need at least version b4361 (Dec. 19th).
But im using ollama not llama.cpp
yeah ollama needs an update to support this, they haven't synced their fork yet
Ollama uses llama.cpp to actually run models. Looks like the latest Ollama release (0.5.4) only has llama.cpp b4327. The development version is currently on b4419, so this should be fixed in the next release.
@Aashraf995 Sorry about this. We're working on a pre-release with the updates, including a llama.cpp update, targeting end of the day for importing external models.
If you run the model using Ollama directly, we've made it work when we released the model:
ollama run phi4
I appreciate the effort, thank you for your hard work .