Issues

• Experiment: lossless weight scheduling for models larger than RAM

  google/gemma.cpp#1036 · Filter by author Mikyx-1 opened on Sep 19, 2026
  2

• Page-size check prevents mmap for blob-aligned checkpoints

  google/gemma.cpp#1031 · Filter by author Mikyx-1 opened on Sep 13, 2026

• Default flash attention retains redundant BF16 KV projection history

  google/gemma.cpp#1029 · Filter by author Mikyx-1 opened on Sep 12, 2026

• Reduce KV memory for local-attention layers

  google/gemma.cpp#1016 · Filter by author Mikyx-1 opened on Sep 6, 2026
  1

• Add experimental W8A8 MatMul with optional QuaRot-style rotation

  google/gemma.cpp#1002 · Filter by author Mikyx-1 opened on Aug 31, 2026
  8

• Feature Request / Proposal: "E12B" (Effective 12B) Model Class for Gemma 5 (Optimized for 12GB VRAM / Desktop Users)

  google/gemma.cpp#990 · Filter by author ozelturktarkan opened on Aug 26, 2026

• rope_theta is ignored on the dense attention path (hardcoded 10000 / 1000000)

  google/gemma.cpp#986 · Filter by author Mikyx-1 opened on Aug 22, 2026
  1

• Add SmolLM2 support — reuse the Qwen3 converter path?

  google/gemma.cpp#982 · Filter by author Mikyx-1 opened on Aug 17, 2026
  5

• Give a known-good example for https://github.com/google/gemma.cpp/blob/main/API_SERVER_README.md

  google/gemma.cpp#911 · Filter by author filmil opened on May 9, 2026
  2

• Support TranslateGemma 4B text-only inference (Gemma 3 4B LM-only, no VIT)

  google/gemma.cpp#888 · Filter by author Selami79 opened on Apr 3, 2026
  4

• Is Gemma4 supported by gemma.cpp?

  google/gemma.cpp#887 · Filter by author YasuakiSakurai opened on Apr 3, 2026
  4

• LLM output regression from #857 building with Emscripten

  google/gemma.cpp#885 · Filter by author brendandahl opened on Mar 26, 2026

• Showcase / question: a board-proven offline language runtime on ESP32-C3, and whether this points to a more extreme form of language/runtime co-design

  google/gemma.cpp#875 · Filter by author Juemian-Deng opened on Mar 18, 2026
  4

• Batch mode in GPU

  google/gemma.cpp#826 · Filter by author uasan opened on Jan 9, 2026
  3

• Segmentation fault using gemma 3 with --image_file candy.ppm

  google/gemma.cpp#817 · Filter by author weiyang121 opened on Dec 26, 2025
  3

• Build fails on Linux (GCC 15): SentencePiece missing <cstdint> + Highway avx10.2-512 errors

  google/gemma.cpp#778 · Filter by author colinrobinsonuib opened on Nov 22, 2025
  3

• integrate gemma.cpp for Android Galaxy s24 devices

  google/gemma.cpp#758 · Filter by author Markus911111 opened on Nov 3, 2025
  3

• Abort / Assertion-error when loading gemma-3/gemmaCpp/3.0-1b-it-sfp

  google/gemma.cpp#726 · Filter by author sschnug opened on Oct 5, 2025
  2

• [User Experience Failure] Impossible to Build and Run on a Resource-Constrained Google Edge Device (Chromebook)

  google/gemma.cpp#688 · Filter by author TaylorShards opened on Sep 3, 2025
  2

• T5Gemma support

  google/gemma.cpp#661 · Filter by author betweenus opened on Aug 16, 2025
  12

• hello_world need to get updated

  google/gemma.cpp#648 · Filter by author KaranocaVe opened on Jul 30, 2025
  2

• Model's reply is strange

  google/gemma.cpp#632 · Filter by author MaeThird opened on Jun 27, 2025
  1

• Getting Error while running paligemma models

  google/gemma.cpp#631 · Filter by author SudarshanShetty1796 opened on Jun 25, 2025
  1

• Segementation fault in prefill phase using Griffin model

  google/gemma.cpp#568 · Filter by author ufownl opened on May 13, 2025
  3

• Optimization: Convert Hidden States to Bfloat16

  google/gemma.cpp#560 · Filter by author FabianSchuetze opened on May 9, 2025
  1