Issues

• OOB read / SEGV via unvalidated gpu_bucket row index in auto-loaded .generated.gpuidx sidecar

  bug-unconfirmed
  Tiiny-AI/PowerInfer#284 · Filter by author 3em0 opened on Sep 24, 2026

• Out-of-bounds read in GGUF model loading: tensor data offsets are not validated against the file size on the default mmap path

  bug-unconfirmed
  Tiiny-AI/PowerInfer#283 · Filter by author 3em0 opened on Sep 24, 2026

• Does TurboQuant offer something tiiny can use?

  question
  Tiiny-AI/PowerInfer#278 · Filter by author tyoc213 opened on Mar 29, 2026

• ChromeOS compatibility and host computer AI computing capability pooling with Tiiny

  question
  Tiiny-AI/PowerInfer#277 · Filter by author HolyPotatoes1 opened on Mar 25, 2026

• Build error ROCm GPU

  bug-unconfirmed
  Tiiny-AI/PowerInfer#276 · Filter by author fmotta opened on Mar 24, 2026
  1

• Any plan on sharing Pocket Lab's available models on hf?

  question
  Tiiny-AI/PowerInfer#274 · Filter by author andreasntr opened on Mar 11, 2026

• Support RK series

  Tiiny-AI/PowerInfer#270 · Filter by author Jay-zzcoder opened on Sep 30, 2025

• Powerinfer's server-side babble has occurred while the command-side looks good

  question
  Tiiny-AI/PowerInfer#269 · Filter by author Chosen-David opened on Aug 19, 2025

• What will SmallThinker do in the prefill stage?

  question
  Tiiny-AI/PowerInfer#268 · Filter by author hebangwen opened on Aug 8, 2025

• Build failure on Linux

  bug-unconfirmed
  Tiiny-AI/PowerInfer#267 · Filter by author Kunde21 opened on Aug 3, 2025
  4

• 有没有什么办法能够让模型对于中文支持地更好一些？

  question
  Tiiny-AI/PowerInfer#265 · Filter by author WineMarks opened on Aug 1, 2025

• Incomplete fallback in libaz for CPUs without AVX2

  bug-unconfirmed
  Tiiny-AI/PowerInfer#264 · Filter by author Yorwba opened on Jul 30, 2025
  3

• UBSAN: runtime error: member access within null pointer of type

  bug-unconfirmed
  Tiiny-AI/PowerInfer#257 · Filter by author JohnSmithBH84 opened on Jun 6, 2025

• Integer Overflow Vulnerability in gguf_fread_str()

  bug-unconfirmed
  Tiiny-AI/PowerInfer#256 · Filter by author Asuk4 opened on Jun 1, 2025

• Predictor training details

  question
  Tiiny-AI/PowerInfer#255 · Filter by author kcwaffle opened on May 13, 2025

• 合作咨询

  Tiiny-AI/PowerInfer#254 · Filter by author dingtine opened on May 12, 2025
  1

• section 6.3.1, computational constraint is not implemented

  question
  Tiiny-AI/PowerInfer#253 · Filter by author jaeyoon-enerzai opened on Apr 27, 2025

• Architecture "SparseLlamaForCausalLM" not supported!

  question
  Tiiny-AI/PowerInfer#252 · Filter by author fxnie opened on Apr 18, 2025

• Your LoRA file

  question
  Tiiny-AI/PowerInfer#251 · Filter by author kimjoohyungsd opened on Apr 16, 2025

• Question on GGUF file

  question
  Tiiny-AI/PowerInfer#250 · Filter by author kimjoohyungsd opened on Apr 15, 2025
  1

• Support for deepseekV3

  enhancement
  Tiiny-AI/PowerInfer#249 · Filter by author sorasoras opened on Mar 29, 2025

• 请问下离线部分预测器的代码有开源吗

  question
  Tiiny-AI/PowerInfer#248 · Filter by author BenetnaschAlkaid opened on Mar 18, 2025

• Fix incorrect sorting of activation_files causing improper FFN offload

  bug-unconfirmed
  Tiiny-AI/PowerInfer#247 · Filter by author prisem123 opened on Mar 11, 2025
  1

• Is it possible to run DeepSeek-R1-Distill-Llama-70B or a 4-bit quantified version on consumer grade computers? This is of great significance

  question
  Tiiny-AI/PowerInfer#244 · Filter by author wusu2016 opened on Feb 13, 2025
  1

• A More User-Friendly Command-Line Interface

  Tiiny-AI/PowerInfer#238 · Filter by author wisman-tccr opened on Jan 25, 2025