Search results

Eval bug: MultiGPU and MultiModels at the same time = 100% GPU

bug-unconfirmed

Status: Open.

#14890 In ggml-org/llama.cpp;

· DKingAlpha opened on Jul 26, 2025

Eval bug: Repeated sequences with gemma3 and image recognition

bug-unconfirmed

Status: Open.

#14888 In ggml-org/llama.cpp;

· deiteris opened on Jul 26, 2025

Eval bug: No kernel named rms_norm_f32_sycl

bug-unconfirmed

Status: Open.

#14887 In ggml-org/llama.cpp;

· FullstackSensei opened on Jul 26, 2025

Eval bug: gemma-3n-E4B-it-Q8_0.gguf is speaking nonsense

bug-unconfirmed

Status: Open.

#14885 In ggml-org/llama.cpp;

· rowheel opened on Jul 26, 2025

Speed regression with -fa and -ctk

bug-unconfirmed

Status: Open.

#14881 In ggml-org/llama.cpp;

· easyfab opened on Jul 25, 2025

Misc. bug:convert_hf_to_gguf.py lora fine-tuning effects for visual recognition are completely lost – only the text-related fine-tuning remains effective.

bug-unconfirmed

Status: Open.

#14867 In ggml-org/llama.cpp;

· YuBin8 opened on Jul 25, 2025

Eval bug: cline plugin for VS Code does not work with any GGUF

bug-unconfirmed

Status: Open.

#14866 In ggml-org/llama.cpp;

· marco-trovato opened on Jul 25, 2025

Misc. bug: slow model loading to GPU when size > 64GB (Vulkan)

bug-unconfirmed

Status: Open.

#14854 In ggml-org/llama.cpp;

· kyuz0 opened on Jul 24, 2025

Eval bug: Embedding output differs significantly between b4712 and b4713

bug-unconfirmed

Status: Open.

#14848 In ggml-org/llama.cpp;

· lingchingteng opened on Jul 24, 2025

Misc. bug: Regression in unified KV cache appears after <code>llama.cpp</code> release b5912 in b5913

bug-unconfirmed

Status: Open.

#14847 In ggml-org/llama.cpp;

· akarasulu opened on Jul 24, 2025

Please implement phi-3-M3-coder

Status: Open.

#14846 In ggml-org/llama.cpp;

· 0wwafa opened on Jul 24, 2025

ggml_vulkan: RADV crash on ggml_set_rows due to zero size buffer

Status: Open.

#14845 In ggml-org/llama.cpp;

· netrunnereve opened on Jul 24, 2025