All pull requests

Pull requests

• fix(i18n): complete Simplified Chinese web UI and add-on translations

  #12941·tzboy08 opened Oct 27, 2026·2 tasks done·

  ·

• fix(faster-whisper): honor language, translate, prompt and temperature

  #12938·fleet2jeep opened Oct 26, 2026·2 tasks done·

  ·

• feat(gallery): add Gemma 4 Artemis E2B and E4B variants

  #12937·Dorijan10 opened Oct 26, 2026·2 tasks done·

  ·

• chore: ⬆️ Update CrispStrobe/CrispASR to 2ed915ae0238fb065943c366e279dd5c11a41a16

  #12933·localai-org-maint-bot opened Oct 26, 2026Collaborator·

  ·

• chore: ⬆️ Update 0xShug0/audio.cpp to 41c1f8ffa0d13a1b7e366e30ea5fe5ab8d169d48

  #12930·localai-org-maint-bot opened Oct 26, 2026Collaborator·

  ·

• feat(gallery): add Qwen3.5 Artemis 2B, 4B and 9B variants

  #12928·Dorijan10 opened Oct 26, 2026·2 tasks done·

  ·

• fix(gallery): pin Stable Audio SFX weights

  #12927·localai-org-maint-bot opened Oct 26, 2026Collaborator·1 of 2 tasks·

  ·
  1

• feat(gallery): add Nanbeige4.3 GGUF variants

  #12926·localai-org-maint-bot opened Oct 26, 2026Collaborator·1 of 2 tasks·

  ·

• chore(model gallery): 🤖 add 0 new models via gallery agent

  #12925·localai-org-maint-bot opened Oct 26, 2026Collaborator·

  ·

• fix(ui): use token usage for chat throughput

  #12923·localai-org-maint-bot opened Oct 26, 2026Collaborator·1 of 2 tasks·

  ·
  1

• feat(gallery): add Qwen 3D Coder preview

  #12922·localai-org-maint-bot opened Oct 26, 2026Collaborator·1 of 2 tasks·

  ·

• [llama-cpp] keep an existing image data URL

  #12921·ilsawa opened Oct 26, 2026·2 tasks done·

  ·
  1

• [llama-cpp] pass reasoning_budget from the model card to the sampler

  #12920·ilsawa opened Oct 26, 2026·2 tasks done·

  ·
  1

• feat(chats): server-side chats with steering, sub-agents and compaction

  #12918·mudler-agent opened Oct 26, 2026Collaborator·1 of 2 tasks·

  ·
  28

• fix(watchdog): track models on address-based external backends

  #12902·bercedev opened Oct 25, 2026Contributor·2 tasks done·

  ·
  1

  1

• feat(qwen3-tts-cpp): batch concurrent requests via qwentts max_batch

  #12897·nexxtmobile opened Oct 25, 2026Contributor·2 tasks done·

  ·
  1

• fix(gallery): render single-brace JSON in Llama 3.1 function examples

  #12891·leilei3167 opened Oct 25, 2026Contributor·2 tasks done·

  ·
  1

  1

• fix(agents): connect agent MCP servers over Streamable HTTP

  #12848·rossigee opened Oct 23, 2026·2 tasks done·

  ·
  1

  1

• chore(deps): bump docs/themes/hugo-theme-relearn from 657eb0c to c7fa39d

  dependenciessubmodules
  #12838·dependabot[bot] opened Oct 22, 2026Bot·

  ·

• chore(deps): bump transformers from 4.48.3 to 5.20.2 in /backend/python/coqui

  dependenciespython
  #12835·dependabot[bot] opened Oct 22, 2026Bot·

  ·

• fix(agentpool): report an empty agent reply as a chat error

  #12824·walcz-de opened Oct 22, 2026Contributor·

  ·
  1

• feat(agentruntime): add runtime foundations and host bridge

  #12808·mudler-agent opened Oct 21, 2026Collaborator·2 tasks done·

  ·
  20

• feat(ui): optionally save chat attachments to a knowledge collection

  #12802·walcz-de opened Oct 21, 2026Contributor·2 tasks done·

  ·
  2

• feat(config): let a profile alias require a tool call and append tool sources

  #12801·walcz-de opened Oct 21, 2026Contributor·2 tasks done·

  ·
  1

• feat(auth): let a legacy owner keep the root storage area

  #12800·walcz-de opened Oct 21, 2026Contributor·2 tasks done·

  ·
  1

  1