Issues

• Fused activation-sparsity + int4 GEMV — happy to contribute the "quantized sparse kernel" from the README

  FasterDecoding/TEAL#16 · Filter by author asmit383 opened on Jul 31, 2026

• Llama-2 proj weight

  FasterDecoding/TEAL#15 · Filter by author shanhx2000 opened on Dec 9, 2025
  2

• Request: Release Training-Based TEAL (Fine-Tuned Version)

  FasterDecoding/TEAL#14 · Filter by author hxu105 opened on Dec 3, 2025

• Incomplete implementation of SparseGEMV

  FasterDecoding/TEAL#12 · Filter by author kyang-06 opened on Sep 2, 2025
  3

• it's strange for the MMLU result

  FasterDecoding/TEAL#10 · Filter by author milktea888 opened on Dec 10, 2024
  1

• When will support Speculative decoding in Training free activation sparsity base code ?

  FasterDecoding/TEAL#5 · Filter by author CarlHuangNuc opened on Sep 18, 2024