All pull requests

Pull requests

• Untrack the assistant model after Gemma4 speculative generate

  Gemma
  #3124·laxmareddyp opened Sep 25, 2026Collaborator·

  ·
  1

  2

• Serialize data_format in ViT, DeiT, ResNet, Xception and EfficientNet

  #3121·GargiGupta-io opened Sep 24, 2026Contributor·

  ·
  1

  3

• Port EditDistance to a pure Python Levenshtein

  #3117·prajak002 opened Sep 21, 2026·

  ·
  1

  3

• Add Muse Glimmer and Muse Glimmer Assistant to Hub

  Gemma
  #3100·ahmed-damda opened Sep 18, 2026·

  ·
  98

• Preserve Llama and Mixtral constructor flags in get_config

  #3099·Anu27n opened Sep 18, 2026·

  ·
  1

  6

• [TF-Free Migration] Make mock Gemma tokenizers work without TensorFlow

  Gemma
  #3097·laxmareddyp opened Sep 18, 2026Collaborator·

  ·
  1

• Add SmolVLM2 model to Hub

  #3096·Ruchitha-dev3 opened Sep 17, 2026Contributor·

  ·
  1

  23

• [TF-Free Migration] Make the test harness importable without TensorFlow

  #3079·laxmareddyp opened Sep 17, 2026Collaborator·

  ·
  1

  36

• Resolve tied weights from safetensors metadata in SafetensorLoader

  #3076·sajithamathi19 opened Sep 16, 2026·

  ·
  1

  5

• Serialize stackwise_nores_option and router_aux_loss_coefficient

  #3073·GargiGupta-io opened Sep 16, 2026Contributor·

  ·
  1

  5

• Add missing converter for vaultgemma

  Gemma
  #3071·dhantule opened Sep 16, 2026Collaborator·

  ·
  1

  4

• Add bloom hf converter

  #3067·pandeyshub-byte opened Sep 13, 2026·

  ·
  1

  5

• Reproduce gpu failure

  #3050·mrinalghoshh opened Sep 10, 2026·

  ·
  2

• Use keras.random.uniform instead of ops.random.uniform in Mixtral dec…

  #3043·tacl202 opened Sep 9, 2026Contributor·

  ·
  1

  1

• Test: do not merge - ignore

  #3036·poornas2505 opened Sep 4, 2026·

  ·
  4

• [OpenVINO] Fix compiled-model cache invalidation and expand test whitelist

  #3029·goyaladitya05 opened Sep 4, 2026Contributor·

  ·
  1

  2

• fix(causal_lm): give the CausalLM.generate_step stub the signature it is called with

  Gemma
  #3028·Anai-Guo opened Sep 3, 2026·

  ·
  1

• [Do not merge ] Remove .kokoro/ directory (migrated to GitHub Actions in #2939)

  #3025·laxmareddyp opened Aug 31, 2026Collaborator·

  ·
  1

• Add Mistral3 model to Hub

  #3020·ahmed-damda opened Aug 27, 2026·

  ·
  74

• feat: add HF safetensors export for qwen3.5_moe

  #2954·mrinalghoshh opened Aug 13, 2026·

  ·
  4

• Add FLUX model to Hub

  new model
  #2944·maitry63 opened Aug 12, 2026Contributor·

  ·
  1

  16

• Serve KerasHub models on vLLM's GPU engine

  Gemma
  #2943·anthony-etim opened Aug 10, 2026Collaborator·

  ·
  21

• Add Qwen3 ASR to Keras-Hub

  #2942·buildwithsuhana opened Aug 8, 2026Contributor·

  ·
  1

  59

• Add KVPress support to KerasHub

  #2941·Ahmed0830 opened Aug 8, 2026Contributor·

  ·
  6

• [Do not Merge] Add GPU test workflow (migrate from Kokoro to GitHub Actions)

  #2939·laxmareddyp opened Aug 8, 2026Collaborator·

  ·
  13