All pull requests

Pull requests

• Xsn/llama batch ext 2

  Apple MetalbuildconversionCUDAdevopsdocumentationggmlHexagonjinja parsermodelmtmdOpenCLOpenVINOserver/uiserverSYCLtestingvendorVulkanWebGPU
  #109·ngxson opened Aug 17, 2026Owner·

  ·
  1

• Xsn/pocket tts

  #108·ngxson opened Aug 5, 2026Owner·

  ·
  11

• Pascal/poc chatterbox

  conversionmtmd
  #107·ngxson opened Aug 2, 2026Owner·

  ·
  49

• Add exaone4 5 (FOR CI)

  examplesmodelpython
  #105·ngxson opened May 26, 2026Owner·

  ·
  1

• (MIRROR) Implement gguf init from buffer

  ggmltesting
  #104·ngxson opened May 11, 2026Owner·

  ·
  4

• Mirror https://github.com/ggml-org/llama.cpp/pull/22121

  examplespythontesting
  #103·ngxson opened Apr 27, 2026Owner·

  ·
  6

• (Mirror) Router: Forward form-data to model server

  examplesserver
  #102·ngxson opened Apr 19, 2026Owner·

  ·
  7

• (Mirror) cli : Use acquire/release semantics for stopping logic

  examplespythonserver/uiserver
  #101·ngxson opened Apr 15, 2026Owner·

  ·
  3

• (WIP) falcon-ocr, for discussion

  examplesmodelpython
  #100·ngxson opened Apr 14, 2026Owner·

  ·
  23

• [Mirror] anthropic: fix prefix caching

  Apple MetalexamplesggmlHexagonNvidia GPUserver
  #98·ngxson opened Apr 12, 2026Owner·

  ·
  12

• fix gguf conversion for audio/vision mmproj (FOR CI)

  examplesmodelpythontesting
  #95·ngxson opened Apr 2, 2026Owner·

  ·
  1

• [Mirror] model : refactor QKV into common build_qkv and create_tensor_qkv helpers

  model
  #94·ngxson opened Apr 1, 2026Owner·

  ·
  9

• wip: server_tools

  examplesserver
  #93·ngxson opened Mar 18, 2026Owner·

  ·
  1

• Xsn/chat fix typed content only (for CI)

  jinja parser
  #87·ngxson opened Feb 9, 2026Owner·

  ·
  1

• Xsn/better tensor debug print (FOR CI)

  model
  #86·ngxson opened Feb 4, 2026Owner·

  ·
  1

• [Mirror] server: /v1/responses (partial)

  examplespythonserver
  #85·ngxson opened Jan 21, 2026Owner·

  ·
  8

• cli : fix reasoning responses in CLI

  examplesservertesting
  #84·ngxson opened Jan 20, 2026Owner·

  ·
  1

• [Mirror] server : refactor oai_parser_opt, move it to server_chat_params

  examplesserver
  #83·ngxson opened Jan 19, 2026Owner·

  ·
  3

• Glm4moelite

  python
  #82·ngxson opened Jan 19, 2026Owner·

  ·
  1

• [Mirror] server: fix memory reservations in populate_token_probs

  examplesserver
  #81·ngxson opened Jan 19, 2026Owner·

  ·
  1

• [Mirror] server: improve slots scheduling for n_cmpl

  examplespythonserver
  #80·ngxson opened Jan 12, 2026Owner·

  ·
  10

• Xsn/remote preset

  documentation
  #79·ngxson opened Jan 8, 2026Owner·

  ·
  1

• vendor : update cpp-httplib to 0.30.0

  examplespythonscriptservertesting
  #78·ngxson opened Jan 7, 2026Owner·

  ·
  1

• server: poc audio gen

  examplesserver
  #77·ngxson opened Jan 7, 2026Owner·

  ·
  1

• Demo: HTTP CORS proxy

  builddevopsexamplesserver
  #76·ngxson opened Jan 6, 2026Owner·

  ·
  2