Search results

WebUI Tauri shell: native desktop wrapper for the llama-server frontend

Status: Open.

#32 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Apr 21, 2026

WebUI: evolve into general-purpose AI workspace with virtual filesystem and artifact rendering

Status: Open.

#31 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Apr 21, 2026

Responses API: misleading error on context overflow, must communicate token limit exceeded

Status: Open.

#19 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 12, 2026

Steering hints: mid-inference context injection via overlapping activations

Status: Open.

#17 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 12, 2026

chat/completions endpoint returns 500 when mmproj is loaded (Qwen3.5-27B VLM)

Status: Open.

#16 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 11, 2026

feat: auto-discover LoRA adapters from models directory

Status: Open.

#14 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 11, 2026

[Frontend] Add lora-adapter toggler / selection list for selected model

Status: Open.

#13 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 11, 2026

Rebase conflict: ht on master (2026-03-11)

sync-conflictUpstream sync rebase conflictUpstream sync rebase conflict

Status: Open.

#12 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 11, 2026

feat: analytical fit for large MoE models (avoid repeated model loads)

Status: Open.

#11 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 7, 2026

Add Native Gemini API Compatibility to Server

Status: Open.

#5 In heiervang-technologies/ht-llama.cpp;

· marksverdhei opened on Mar 6, 2026